Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+iMF-XL-2/demo.png filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,72 @@
+---
+license: mit
+library_name: diffusers
+pipeline_tag: text-to-image
+tags:
+- diffusers
+- imf
+- image-generation
+- class-conditional
+- imagenet
+inference: true
+widget:
+- output:
+    url: iMF-XL-2/demo.png
+language:
+- en
+---
+
+# iMF-diffusers
+
+Native diffusers implementation of [Improved Mean Flows (iMF)](https://arxiv.org/abs/2512.02012). Each variant folder is self-contained:
+
+- `pipeline.py` — `IMFPipeline`
+- `scheduler/scheduler_config.json` — `FlowMatchEulerDiscreteScheduler` config
+- `transformer/transformer_imf.py` — `IMFTransformer2DModel`
+- `vae/` — bundled `stabilityai/sd-vae-ft-mse` (`AutoencoderKL`)
+
+## Demo
+
+![iMF-XL-2 demo](iMF-XL-2/demo.png)
+
+Class-conditional sample (ImageNet class **207**, golden retriever), `iMF-XL/2` at 256×256, 1 step, CFG 1.8, interval [0.0, 1.0], seed 42.
+
+## Available checkpoints
+
+| Checkpoint | Path | Latent size | FID eval CFG (ω) | FID eval interval |
+| --- | --- | --- | --- | --- |
+| iMF-B/2 | `./iMF-B-2` | 32×32 | 8.0 | [0.40, 0.65] |
+| iMF-L/2 | `./iMF-L-2` | 32×32 | 10.5 | [0.40, 0.60] |
+| iMF-XL/2 | `./iMF-XL-2` | 32×32 | 8.0 | [0.42, 0.62] |
+
+FID eval settings follow upstream [imeanflow eval config](https://github.com/Lyy-iiis/imeanflow/blob/main/configs/eval_config.yml).
+
+## Inference
+
+```python
+from pathlib import Path
+from diffusers import DiffusionPipeline
+import torch
+
+model_dir = Path("./iMF-XL-2")
+pipe = DiffusionPipeline.from_pretrained(
+    str(model_dir),
+    local_files_only=True,
+    custom_pipeline=str(model_dir / "pipeline.py"),
+    trust_remote_code=True,
+    torch_dtype=torch.bfloat16,
+).to("cuda")
+
+generator = torch.Generator(device="cuda").manual_seed(42)
+image = pipe(
+    class_labels="golden retriever",
+    num_inference_steps=1,
+    guidance_scale=1.8,
+    guidance_interval_start=0.0,
+    guidance_interval_end=1.0,
+    generator=generator,
+).images[0]
+image.save("demo.png")
+```
+
+Load a **variant subfolder** (e.g. `./iMF-XL-2`), not the repo root.

iMF-B-2/README.md

@@ -0,0 +1,40 @@
+---
+license: mit
+library_name: diffusers
+pipeline_tag: text-to-image
+tags:
+- diffusers
+- imf
+- image-generation
+- class-conditional
+---
+
+# iMF-B-2
+
+Self-contained Diffusers variant for **iMF-B/2** (Improved Mean Flows).
+
+## Load
+
+```python
+from pathlib import Path
+from diffusers import DiffusionPipeline
+import torch
+
+model_dir = Path("iMF-B-2")
+pipe = DiffusionPipeline.from_pretrained(
+    str(model_dir),
+    local_files_only=True,
+    custom_pipeline=str(model_dir / "pipeline.py"),
+    trust_remote_code=True,
+    torch_dtype=torch.float32,
+)
+pipe.to("cuda")
+
+image = pipe(
+    class_labels=207,
+    num_inference_steps=1,
+    guidance_scale=2.7,
+    guidance_interval_start=0.1,
+    guidance_interval_end=0.9,
+).images[0]
+```

iMF-B-2/model_index.json

@@ -0,0 +1,1021 @@
+{
+  "_class_name": [
+    "pipeline",
+    "IMFPipeline"
+  ],
+  "_diffusers_version": "0.36.0",
+  "scheduler": [
+    "diffusers",
+    "FlowMatchEulerDiscreteScheduler"
+  ],
+  "transformer": [
+    "transformer_imf",
+    "IMFTransformer2DModel"
+  ],
+  "vae": [
+    "diffusers",
+    "AutoencoderKL"
+  ],
+  "id2label": {
+    "0": "tench, Tinca tinca",
+    "1": "goldfish, Carassius auratus",
+    "2": "great white shark, white shark, man-eater, man-eating shark, Carcharodon carcharias",
+    "3": "tiger shark, Galeocerdo cuvieri",
+    "4": "hammerhead, hammerhead shark",
+    "5": "electric ray, crampfish, numbfish, torpedo",
+    "6": "stingray",
+    "7": "cock",
+    "8": "hen",
+    "9": "ostrich, Struthio camelus",
+    "10": "brambling, Fringilla montifringilla",
+    "11": "goldfinch, Carduelis carduelis",
+    "12": "house finch, linnet, Carpodacus mexicanus",
+    "13": "junco, snowbird",
+    "14": "indigo bunting, indigo finch, indigo bird, Passerina cyanea",
+    "15": "robin, American robin, Turdus migratorius",
+    "16": "bulbul",
+    "17": "jay",
+    "18": "magpie",
+    "19": "chickadee",
+    "20": "water ouzel, dipper",
+    "21": "kite",
+    "22": "bald eagle, American eagle, Haliaeetus leucocephalus",
+    "23": "vulture",
+    "24": "great grey owl, great gray owl, Strix nebulosa",
+    "25": "European fire salamander, Salamandra salamandra",
+    "26": "common newt, Triturus vulgaris",
+    "27": "eft",
+    "28": "spotted salamander, Ambystoma maculatum",
+    "29": "axolotl, mud puppy, Ambystoma mexicanum",
+    "30": "bullfrog, Rana catesbeiana",
+    "31": "tree frog, tree-frog",
+    "32": "tailed frog, bell toad, ribbed toad, tailed toad, Ascaphus trui",
+    "33": "loggerhead, loggerhead turtle, Caretta caretta",
+    "34": "leatherback turtle, leatherback, leathery turtle, Dermochelys coriacea",
+    "35": "mud turtle",
+    "36": "terrapin",
+    "37": "box turtle, box tortoise",
+    "38": "banded gecko",
+    "39": "common iguana, iguana, Iguana iguana",
+    "40": "American chameleon, anole, Anolis carolinensis",
+    "41": "whiptail, whiptail lizard",
+    "42": "agama",
+    "43": "frilled lizard, Chlamydosaurus kingi",
+    "44": "alligator lizard",
+    "45": "Gila monster, Heloderma suspectum",
+    "46": "green lizard, Lacerta viridis",
+    "47": "African chameleon, Chamaeleo chamaeleon",
+    "48": "Komodo dragon, Komodo lizard, dragon lizard, giant lizard, Varanus komodoensis",
+    "49": "African crocodile, Nile crocodile, Crocodylus niloticus",
+    "50": "American alligator, Alligator mississipiensis",
+    "51": "triceratops",
+    "52": "thunder snake, worm snake, Carphophis amoenus",
+    "53": "ringneck snake, ring-necked snake, ring snake",
+    "54": "hognose snake, puff adder, sand viper",
+    "55": "green snake, grass snake",
+    "56": "king snake, kingsnake",
+    "57": "garter snake, grass snake",
+    "58": "water snake",
+    "59": "vine snake",
+    "60": "night snake, Hypsiglena torquata",
+    "61": "boa constrictor, Constrictor constrictor",
+    "62": "rock python, rock snake, Python sebae",
+    "63": "Indian cobra, Naja naja",
+    "64": "green mamba",
+    "65": "sea snake",
+    "66": "horned viper, cerastes, sand viper, horned asp, Cerastes cornutus",
+    "67": "diamondback, diamondback rattlesnake, Crotalus adamanteus",
+    "68": "sidewinder, horned rattlesnake, Crotalus cerastes",
+    "69": "trilobite",
+    "70": "harvestman, daddy longlegs, Phalangium opilio",
+    "71": "scorpion",
+    "72": "black and gold garden spider, Argiope aurantia",
+    "73": "barn spider, Araneus cavaticus",
+    "74": "garden spider, Aranea diademata",
+    "75": "black widow, Latrodectus mactans",
+    "76": "tarantula",
+    "77": "wolf spider, hunting spider",
+    "78": "tick",
+    "79": "centipede",
+    "80": "black grouse",
+    "81": "ptarmigan",
+    "82": "ruffed grouse, partridge, Bonasa umbellus",
+    "83": "prairie chicken, prairie grouse, prairie fowl",
+    "84": "peacock",
+    "85": "quail",
+    "86": "partridge",
+    "87": "African grey, African gray, Psittacus erithacus",
+    "88": "macaw",
+    "89": "sulphur-crested cockatoo, Kakatoe galerita, Cacatua galerita",
+    "90": "lorikeet",
+    "91": "coucal",
+    "92": "bee eater",
+    "93": "hornbill",
+    "94": "hummingbird",
+    "95": "jacamar",
+    "96": "toucan",
+    "97": "drake",
+    "98": "red-breasted merganser, Mergus serrator",
+    "99": "goose",
+    "100": "black swan, Cygnus atratus",
+    "101": "tusker",
+    "102": "echidna, spiny anteater, anteater",
+    "103": "platypus, duckbill, duckbilled platypus, duck-billed platypus, Ornithorhynchus anatinus",
+    "104": "wallaby, brush kangaroo",
+    "105": "koala, koala bear, kangaroo bear, native bear, Phascolarctos cinereus",
+    "106": "wombat",
+    "107": "jellyfish",
+    "108": "sea anemone, anemone",
+    "109": "brain coral",
+    "110": "flatworm, platyhelminth",
+    "111": "nematode, nematode worm, roundworm",
+    "112": "conch",
+    "113": "snail",
+    "114": "slug",
+    "115": "sea slug, nudibranch",
+    "116": "chiton, coat-of-mail shell, sea cradle, polyplacophore",
+    "117": "chambered nautilus, pearly nautilus, nautilus",
+    "118": "Dungeness crab, Cancer magister",
+    "119": "rock crab, Cancer irroratus",
+    "120": "fiddler crab",
+    "121": "king crab, Alaska crab, Alaskan king crab, Alaska king crab, Paralithodes camtschatica",
+    "122": "American lobster, Northern lobster, Maine lobster, Homarus americanus",
+    "123": "spiny lobster, langouste, rock lobster, crawfish, crayfish, sea crawfish",
+    "124": "crayfish, crawfish, crawdad, crawdaddy",
+    "125": "hermit crab",
+    "126": "isopod",
+    "127": "white stork, Ciconia ciconia",
+    "128": "black stork, Ciconia nigra",
+    "129": "spoonbill",
+    "130": "flamingo",
+    "131": "little blue heron, Egretta caerulea",
+    "132": "American egret, great white heron, Egretta albus",
+    "133": "bittern",
+    "134": "crane",
+    "135": "limpkin, Aramus pictus",
+    "136": "European gallinule, Porphyrio porphyrio",
+    "137": "American coot, marsh hen, mud hen, water hen, Fulica americana",
+    "138": "bustard",
+    "139": "ruddy turnstone, Arenaria interpres",
+    "140": "red-backed sandpiper, dunlin, Erolia alpina",
+    "141": "redshank, Tringa totanus",
+    "142": "dowitcher",
+    "143": "oystercatcher, oyster catcher",
+    "144": "pelican",
+    "145": "king penguin, Aptenodytes patagonica",
+    "146": "albatross, mollymawk",
+    "147": "grey whale, gray whale, devilfish, Eschrichtius gibbosus, Eschrichtius robustus",
+    "148": "killer whale, killer, orca, grampus, sea wolf, Orcinus orca",
+    "149": "dugong, Dugong dugon",
+    "150": "sea lion",
+    "151": "Chihuahua",
+    "152": "Japanese spaniel",
+    "153": "Maltese dog, Maltese terrier, Maltese",
+    "154": "Pekinese, Pekingese, Peke",
+    "155": "Shih-Tzu",
+    "156": "Blenheim spaniel",
+    "157": "papillon",
+    "158": "toy terrier",
+    "159": "Rhodesian ridgeback",
+    "160": "Afghan hound, Afghan",
+    "161": "basset, basset hound",
+    "162": "beagle",
+    "163": "bloodhound, sleuthhound",
+    "164": "bluetick",
+    "165": "black-and-tan coonhound",
+    "166": "Walker hound, Walker foxhound",
+    "167": "English foxhound",
+    "168": "redbone",
+    "169": "borzoi, Russian wolfhound",
+    "170": "Irish wolfhound",
+    "171": "Italian greyhound",
+    "172": "whippet",
+    "173": "Ibizan hound, Ibizan Podenco",
+    "174": "Norwegian elkhound, elkhound",
+    "175": "otterhound, otter hound",
+    "176": "Saluki, gazelle hound",
+    "177": "Scottish deerhound, deerhound",
+    "178": "Weimaraner",
+    "179": "Staffordshire bullterrier, Staffordshire bull terrier",
+    "180": "American Staffordshire terrier, Staffordshire terrier, American pit bull terrier, pit bull terrier",
+    "181": "Bedlington terrier",
+    "182": "Border terrier",
+    "183": "Kerry blue terrier",
+    "184": "Irish terrier",
+    "185": "Norfolk terrier",
+    "186": "Norwich terrier",
+    "187": "Yorkshire terrier",
+    "188": "wire-haired fox terrier",
+    "189": "Lakeland terrier",
+    "190": "Sealyham terrier, Sealyham",
+    "191": "Airedale, Airedale terrier",
+    "192": "cairn, cairn terrier",
+    "193": "Australian terrier",
+    "194": "Dandie Dinmont, Dandie Dinmont terrier",
+    "195": "Boston bull, Boston terrier",
+    "196": "miniature schnauzer",
+    "197": "giant schnauzer",
+    "198": "standard schnauzer",
+    "199": "Scotch terrier, Scottish terrier, Scottie",
+    "200": "Tibetan terrier, chrysanthemum dog",
+    "201": "silky terrier, Sydney silky",
+    "202": "soft-coated wheaten terrier",
+    "203": "West Highland white terrier",
+    "204": "Lhasa, Lhasa apso",
+    "205": "flat-coated retriever",
+    "206": "curly-coated retriever",
+    "207": "golden retriever",
+    "208": "Labrador retriever",
+    "209": "Chesapeake Bay retriever",
+    "210": "German short-haired pointer",
+    "211": "vizsla, Hungarian pointer",
+    "212": "English setter",
+    "213": "Irish setter, red setter",
+    "214": "Gordon setter",
+    "215": "Brittany spaniel",
+    "216": "clumber, clumber spaniel",
+    "217": "English springer, English springer spaniel",
+    "218": "Welsh springer spaniel",
+    "219": "cocker spaniel, English cocker spaniel, cocker",
+    "220": "Sussex spaniel",
+    "221": "Irish water spaniel",
+    "222": "kuvasz",
+    "223": "schipperke",
+    "224": "groenendael",
+    "225": "malinois",
+    "226": "briard",
+    "227": "kelpie",
+    "228": "komondor",
+    "229": "Old English sheepdog, bobtail",
+    "230": "Shetland sheepdog, Shetland sheep dog, Shetland",
+    "231": "collie",
+    "232": "Border collie",
+    "233": "Bouvier des Flandres, Bouviers des Flandres",
+    "234": "Rottweiler",
+    "235": "German shepherd, German shepherd dog, German police dog, alsatian",
+    "236": "Doberman, Doberman pinscher",
+    "237": "miniature pinscher",
+    "238": "Greater Swiss Mountain dog",
+    "239": "Bernese mountain dog",
+    "240": "Appenzeller",
+    "241": "EntleBucher",
+    "242": "boxer",
+    "243": "bull mastiff",
+    "244": "Tibetan mastiff",
+    "245": "French bulldog",
+    "246": "Great Dane",
+    "247": "Saint Bernard, St Bernard",
+    "248": "Eskimo dog, husky",
+    "249": "malamute, malemute, Alaskan malamute",
+    "250": "Siberian husky",
+    "251": "dalmatian, coach dog, carriage dog",
+    "252": "affenpinscher, monkey pinscher, monkey dog",
+    "253": "basenji",
+    "254": "pug, pug-dog",
+    "255": "Leonberg",
+    "256": "Newfoundland, Newfoundland dog",
+    "257": "Great Pyrenees",
+    "258": "Samoyed, Samoyede",
+    "259": "Pomeranian",
+    "260": "chow, chow chow",
+    "261": "keeshond",
+    "262": "Brabancon griffon",
+    "263": "Pembroke, Pembroke Welsh corgi",
+    "264": "Cardigan, Cardigan Welsh corgi",
+    "265": "toy poodle",
+    "266": "miniature poodle",
+    "267": "standard poodle",
+    "268": "Mexican hairless",
+    "269": "timber wolf, grey wolf, gray wolf, Canis lupus",
+    "270": "white wolf, Arctic wolf, Canis lupus tundrarum",
+    "271": "red wolf, maned wolf, Canis rufus, Canis niger",
+    "272": "coyote, prairie wolf, brush wolf, Canis latrans",
+    "273": "dingo, warrigal, warragal, Canis dingo",
+    "274": "dhole, Cuon alpinus",
+    "275": "African hunting dog, hyena dog, Cape hunting dog, Lycaon pictus",
+    "276": "hyena, hyaena",
+    "277": "red fox, Vulpes vulpes",
+    "278": "kit fox, Vulpes macrotis",
+    "279": "Arctic fox, white fox, Alopex lagopus",
+    "280": "grey fox, gray fox, Urocyon cinereoargenteus",
+    "281": "tabby, tabby cat",
+    "282": "tiger cat",
+    "283": "Persian cat",
+    "284": "Siamese cat, Siamese",
+    "285": "Egyptian cat",
+    "286": "cougar, puma, catamount, mountain lion, painter, panther, Felis concolor",
+    "287": "lynx, catamount",
+    "288": "leopard, Panthera pardus",
+    "289": "snow leopard, ounce, Panthera uncia",
+    "290": "jaguar, panther, Panthera onca, Felis onca",
+    "291": "lion, king of beasts, Panthera leo",
+    "292": "tiger, Panthera tigris",
+    "293": "cheetah, chetah, Acinonyx jubatus",
+    "294": "brown bear, bruin, Ursus arctos",
+    "295": "American black bear, black bear, Ursus americanus, Euarctos americanus",
+    "296": "ice bear, polar bear, Ursus Maritimus, Thalarctos maritimus",
+    "297": "sloth bear, Melursus ursinus, Ursus ursinus",
+    "298": "mongoose",
+    "299": "meerkat, mierkat",
+    "300": "tiger beetle",
+    "301": "ladybug, ladybeetle, lady beetle, ladybird, ladybird beetle",
+    "302": "ground beetle, carabid beetle",
+    "303": "long-horned beetle, longicorn, longicorn beetle",
+    "304": "leaf beetle, chrysomelid",
+    "305": "dung beetle",
+    "306": "rhinoceros beetle",
+    "307": "weevil",
+    "308": "fly",
+    "309": "bee",
+    "310": "ant, emmet, pismire",
+    "311": "grasshopper, hopper",
+    "312": "cricket",
+    "313": "walking stick, walkingstick, stick insect",
+    "314": "cockroach, roach",
+    "315": "mantis, mantid",
+    "316": "cicada, cicala",
+    "317": "leafhopper",
+    "318": "lacewing, lacewing fly",
+    "319": "dragonfly, darning needle, devils darning needle, sewing needle, snake feeder, snake doctor, mosquito hawk, skeeter hawk",
+    "320": "damselfly",
+    "321": "admiral",
+    "322": "ringlet, ringlet butterfly",
+    "323": "monarch, monarch butterfly, milkweed butterfly, Danaus plexippus",
+    "324": "cabbage butterfly",
+    "325": "sulphur butterfly, sulfur butterfly",
+    "326": "lycaenid, lycaenid butterfly",
+    "327": "starfish, sea star",
+    "328": "sea urchin",
+    "329": "sea cucumber, holothurian",
+    "330": "wood rabbit, cottontail, cottontail rabbit",
+    "331": "hare",
+    "332": "Angora, Angora rabbit",
+    "333": "hamster",
+    "334": "porcupine, hedgehog",
+    "335": "fox squirrel, eastern fox squirrel, Sciurus niger",
+    "336": "marmot",
+    "337": "beaver",
+    "338": "guinea pig, Cavia cobaya",
+    "339": "sorrel",
+    "340": "zebra",
+    "341": "hog, pig, grunter, squealer, Sus scrofa",
+    "342": "wild boar, boar, Sus scrofa",
+    "343": "warthog",
+    "344": "hippopotamus, hippo, river horse, Hippopotamus amphibius",
+    "345": "ox",
+    "346": "water buffalo, water ox, Asiatic buffalo, Bubalus bubalis",
+    "347": "bison",
+    "348": "ram, tup",
+    "349": "bighorn, bighorn sheep, cimarron, Rocky Mountain bighorn, Rocky Mountain sheep, Ovis canadensis",
+    "350": "ibex, Capra ibex",
+    "351": "hartebeest",
+    "352": "impala, Aepyceros melampus",
+    "353": "gazelle",
+    "354": "Arabian camel, dromedary, Camelus dromedarius",
+    "355": "llama",
+    "356": "weasel",
+    "357": "mink",
+    "358": "polecat, fitch, foulmart, foumart, Mustela putorius",
+    "359": "black-footed ferret, ferret, Mustela nigripes",
+    "360": "otter",
+    "361": "skunk, polecat, wood pussy",
+    "362": "badger",
+    "363": "armadillo",
+    "364": "three-toed sloth, ai, Bradypus tridactylus",
+    "365": "orangutan, orang, orangutang, Pongo pygmaeus",
+    "366": "gorilla, Gorilla gorilla",
+    "367": "chimpanzee, chimp, Pan troglodytes",
+    "368": "gibbon, Hylobates lar",
+    "369": "siamang, Hylobates syndactylus, Symphalangus syndactylus",
+    "370": "guenon, guenon monkey",
+    "371": "patas, hussar monkey, Erythrocebus patas",
+    "372": "baboon",
+    "373": "macaque",
+    "374": "langur",
+    "375": "colobus, colobus monkey",
+    "376": "proboscis monkey, Nasalis larvatus",
+    "377": "marmoset",
+    "378": "capuchin, ringtail, Cebus capucinus",
+    "379": "howler monkey, howler",
+    "380": "titi, titi monkey",
+    "381": "spider monkey, Ateles geoffroyi",
+    "382": "squirrel monkey, Saimiri sciureus",
+    "383": "Madagascar cat, ring-tailed lemur, Lemur catta",
+    "384": "indri, indris, Indri indri, Indri brevicaudatus",
+    "385": "Indian elephant, Elephas maximus",
+    "386": "African elephant, Loxodonta africana",
+    "387": "lesser panda, red panda, panda, bear cat, cat bear, Ailurus fulgens",
+    "388": "giant panda, panda, panda bear, coon bear, Ailuropoda melanoleuca",
+    "389": "barracouta, snoek",
+    "390": "eel",
+    "391": "coho, cohoe, coho salmon, blue jack, silver salmon, Oncorhynchus kisutch",
+    "392": "rock beauty, Holocanthus tricolor",
+    "393": "anemone fish",
+    "394": "sturgeon",
+    "395": "gar, garfish, garpike, billfish, Lepisosteus osseus",
+    "396": "lionfish",
+    "397": "puffer, pufferfish, blowfish, globefish",
+    "398": "abacus",
+    "399": "abaya",
+    "400": "academic gown, academic robe, judge robe",
+    "401": "accordion, piano accordion, squeeze box",
+    "402": "acoustic guitar",
+    "403": "aircraft carrier, carrier, flattop, attack aircraft carrier",
+    "404": "airliner",
+    "405": "airship, dirigible",
+    "406": "altar",
+    "407": "ambulance",
+    "408": "amphibian, amphibious vehicle",
+    "409": "analog clock",
+    "410": "apiary, bee house",
+    "411": "apron",
+    "412": "ashcan, trash can, garbage can, wastebin, ash bin, ash-bin, ashbin, dustbin, trash barrel, trash bin",
+    "413": "assault rifle, assault gun",
+    "414": "backpack, back pack, knapsack, packsack, rucksack, haversack",
+    "415": "bakery, bakeshop, bakehouse",
+    "416": "balance beam, beam",
+    "417": "balloon",
+    "418": "ballpoint, ballpoint pen, ballpen, Biro",
+    "419": "Band Aid",
+    "420": "banjo",
+    "421": "bannister, banister, balustrade, balusters, handrail",
+    "422": "barbell",
+    "423": "barber chair",
+    "424": "barbershop",
+    "425": "barn",
+    "426": "barometer",
+    "427": "barrel, cask",
+    "428": "barrow, garden cart, lawn cart, wheelbarrow",
+    "429": "baseball",
+    "430": "basketball",
+    "431": "bassinet",
+    "432": "bassoon",
+    "433": "bathing cap, swimming cap",
+    "434": "bath towel",
+    "435": "bathtub, bathing tub, bath, tub",
+    "436": "beach wagon, station wagon, wagon, estate car, beach waggon, station waggon, waggon",
+    "437": "beacon, lighthouse, beacon light, pharos",
+    "438": "beaker",
+    "439": "bearskin, busby, shako",
+    "440": "beer bottle",
+    "441": "beer glass",
+    "442": "bell cote, bell cot",
+    "443": "bib",
+    "444": "bicycle-built-for-two, tandem bicycle, tandem",
+    "445": "bikini, two-piece",
+    "446": "binder, ring-binder",
+    "447": "binoculars, field glasses, opera glasses",
+    "448": "birdhouse",
+    "449": "boathouse",
+    "450": "bobsled, bobsleigh, bob",
+    "451": "bolo tie, bolo, bola tie, bola",
+    "452": "bonnet, poke bonnet",
+    "453": "bookcase",
+    "454": "bookshop, bookstore, bookstall",
+    "455": "bottlecap",
+    "456": "bow",
+    "457": "bow tie, bow-tie, bowtie",
+    "458": "brass, memorial tablet, plaque",
+    "459": "brassiere, bra, bandeau",
+    "460": "breakwater, groin, groyne, mole, bulwark, seawall, jetty",
+    "461": "breastplate, aegis, egis",
+    "462": "broom",
+    "463": "bucket, pail",
+    "464": "buckle",
+    "465": "bulletproof vest",
+    "466": "bullet train, bullet",
+    "467": "butcher shop, meat market",
+    "468": "cab, hack, taxi, taxicab",
+    "469": "caldron, cauldron",
+    "470": "candle, taper, wax light",
+    "471": "cannon",
+    "472": "canoe",
+    "473": "can opener, tin opener",
+    "474": "cardigan",
+    "475": "car mirror",
+    "476": "carousel, carrousel, merry-go-round, roundabout, whirligig",
+    "477": "carpenters kit, tool kit",
+    "478": "carton",
+    "479": "car wheel",
+    "480": "cash machine, cash dispenser, automated teller machine, automatic teller machine, automated teller, automatic teller, ATM",
+    "481": "cassette",
+    "482": "cassette player",
+    "483": "castle",
+    "484": "catamaran",
+    "485": "CD player",
+    "486": "cello, violoncello",
+    "487": "cellular telephone, cellular phone, cellphone, cell, mobile phone",
+    "488": "chain",
+    "489": "chainlink fence",
+    "490": "chain mail, ring mail, mail, chain armor, chain armour, ring armor, ring armour",
+    "491": "chain saw, chainsaw",
+    "492": "chest",
+    "493": "chiffonier, commode",
+    "494": "chime, bell, gong",
+    "495": "china cabinet, china closet",
+    "496": "Christmas stocking",
+    "497": "church, church building",
+    "498": "cinema, movie theater, movie theatre, movie house, picture palace",
+    "499": "cleaver, meat cleaver, chopper",
+    "500": "cliff dwelling",
+    "501": "cloak",
+    "502": "clog, geta, patten, sabot",
+    "503": "cocktail shaker",
+    "504": "coffee mug",
+    "505": "coffeepot",
+    "506": "coil, spiral, volute, whorl, helix",
+    "507": "combination lock",
+    "508": "computer keyboard, keypad",
+    "509": "confectionery, confectionary, candy store",
+    "510": "container ship, containership, container vessel",
+    "511": "convertible",
+    "512": "corkscrew, bottle screw",
+    "513": "cornet, horn, trumpet, trump",
+    "514": "cowboy boot",
+    "515": "cowboy hat, ten-gallon hat",
+    "516": "cradle",
+    "517": "crane",
+    "518": "crash helmet",
+    "519": "crate",
+    "520": "crib, cot",
+    "521": "Crock Pot",
+    "522": "croquet ball",
+    "523": "crutch",
+    "524": "cuirass",
+    "525": "dam, dike, dyke",
+    "526": "desk",
+    "527": "desktop computer",
+    "528": "dial telephone, dial phone",
+    "529": "diaper, nappy, napkin",
+    "530": "digital clock",
+    "531": "digital watch",
+    "532": "dining table, board",
+    "533": "dishrag, dishcloth",
+    "534": "dishwasher, dish washer, dishwashing machine",
+    "535": "disk brake, disc brake",
+    "536": "dock, dockage, docking facility",
+    "537": "dogsled, dog sled, dog sleigh",
+    "538": "dome",
+    "539": "doormat, welcome mat",
+    "540": "drilling platform, offshore rig",
+    "541": "drum, membranophone, tympan",
+    "542": "drumstick",
+    "543": "dumbbell",
+    "544": "Dutch oven",
+    "545": "electric fan, blower",
+    "546": "electric guitar",
+    "547": "electric locomotive",
+    "548": "entertainment center",
+    "549": "envelope",
+    "550": "espresso maker",
+    "551": "face powder",
+    "552": "feather boa, boa",
+    "553": "file, file cabinet, filing cabinet",
+    "554": "fireboat",
+    "555": "fire engine, fire truck",
+    "556": "fire screen, fireguard",
+    "557": "flagpole, flagstaff",
+    "558": "flute, transverse flute",
+    "559": "folding chair",
+    "560": "football helmet",
+    "561": "forklift",
+    "562": "fountain",
+    "563": "fountain pen",
+    "564": "four-poster",
+    "565": "freight car",
+    "566": "French horn, horn",
+    "567": "frying pan, frypan, skillet",
+    "568": "fur coat",
+    "569": "garbage truck, dustcart",
+    "570": "gasmask, respirator, gas helmet",
+    "571": "gas pump, gasoline pump, petrol pump, island dispenser",
+    "572": "goblet",
+    "573": "go-kart",
+    "574": "golf ball",
+    "575": "golfcart, golf cart",
+    "576": "gondola",
+    "577": "gong, tam-tam",
+    "578": "gown",
+    "579": "grand piano, grand",
+    "580": "greenhouse, nursery, glasshouse",
+    "581": "grille, radiator grille",
+    "582": "grocery store, grocery, food market, market",
+    "583": "guillotine",
+    "584": "hair slide",
+    "585": "hair spray",
+    "586": "half track",
+    "587": "hammer",
+    "588": "hamper",
+    "589": "hand blower, blow dryer, blow drier, hair dryer, hair drier",
+    "590": "hand-held computer, hand-held microcomputer",
+    "591": "handkerchief, hankie, hanky, hankey",
+    "592": "hard disc, hard disk, fixed disk",
+    "593": "harmonica, mouth organ, harp, mouth harp",
+    "594": "harp",
+    "595": "harvester, reaper",
+    "596": "hatchet",
+    "597": "holster",
+    "598": "home theater, home theatre",
+    "599": "honeycomb",
+    "600": "hook, claw",
+    "601": "hoopskirt, crinoline",
+    "602": "horizontal bar, high bar",
+    "603": "horse cart, horse-cart",
+    "604": "hourglass",
+    "605": "iPod",
+    "606": "iron, smoothing iron",
+    "607": "jack-o-lantern",
+    "608": "jean, blue jean, denim",
+    "609": "jeep, landrover",
+    "610": "jersey, T-shirt, tee shirt",
+    "611": "jigsaw puzzle",
+    "612": "jinrikisha, ricksha, rickshaw",
+    "613": "joystick",
+    "614": "kimono",
+    "615": "knee pad",
+    "616": "knot",
+    "617": "lab coat, laboratory coat",
+    "618": "ladle",
+    "619": "lampshade, lamp shade",
+    "620": "laptop, laptop computer",
+    "621": "lawn mower, mower",
+    "622": "lens cap, lens cover",
+    "623": "letter opener, paper knife, paperknife",
+    "624": "library",
+    "625": "lifeboat",
+    "626": "lighter, light, igniter, ignitor",
+    "627": "limousine, limo",
+    "628": "liner, ocean liner",
+    "629": "lipstick, lip rouge",
+    "630": "Loafer",
+    "631": "lotion",
+    "632": "loudspeaker, speaker, speaker unit, loudspeaker system, speaker system",
+    "633": "loupe, jewelers loupe",
+    "634": "lumbermill, sawmill",
+    "635": "magnetic compass",
+    "636": "mailbag, postbag",
+    "637": "mailbox, letter box",
+    "638": "maillot",
+    "639": "maillot, tank suit",
+    "640": "manhole cover",
+    "641": "maraca",
+    "642": "marimba, xylophone",
+    "643": "mask",
+    "644": "matchstick",
+    "645": "maypole",
+    "646": "maze, labyrinth",
+    "647": "measuring cup",
+    "648": "medicine chest, medicine cabinet",
+    "649": "megalith, megalithic structure",
+    "650": "microphone, mike",
+    "651": "microwave, microwave oven",
+    "652": "military uniform",
+    "653": "milk can",
+    "654": "minibus",
+    "655": "miniskirt, mini",
+    "656": "minivan",
+    "657": "missile",
+    "658": "mitten",
+    "659": "mixing bowl",
+    "660": "mobile home, manufactured home",
+    "661": "Model T",
+    "662": "modem",
+    "663": "monastery",
+    "664": "monitor",
+    "665": "moped",
+    "666": "mortar",
+    "667": "mortarboard",
+    "668": "mosque",
+    "669": "mosquito net",
+    "670": "motor scooter, scooter",
+    "671": "mountain bike, all-terrain bike, off-roader",
+    "672": "mountain tent",
+    "673": "mouse, computer mouse",
+    "674": "mousetrap",
+    "675": "moving van",
+    "676": "muzzle",
+    "677": "nail",
+    "678": "neck brace",
+    "679": "necklace",
+    "680": "nipple",
+    "681": "notebook, notebook computer",
+    "682": "obelisk",
+    "683": "oboe, hautboy, hautbois",
+    "684": "ocarina, sweet potato",
+    "685": "odometer, hodometer, mileometer, milometer",
+    "686": "oil filter",
+    "687": "organ, pipe organ",
+    "688": "oscilloscope, scope, cathode-ray oscilloscope, CRO",
+    "689": "overskirt",
+    "690": "oxcart",
+    "691": "oxygen mask",
+    "692": "packet",
+    "693": "paddle, boat paddle",
+    "694": "paddlewheel, paddle wheel",
+    "695": "padlock",
+    "696": "paintbrush",
+    "697": "pajama, pyjama, pjs, jammies",
+    "698": "palace",
+    "699": "panpipe, pandean pipe, syrinx",
+    "700": "paper towel",
+    "701": "parachute, chute",
+    "702": "parallel bars, bars",
+    "703": "park bench",
+    "704": "parking meter",
+    "705": "passenger car, coach, carriage",
+    "706": "patio, terrace",
+    "707": "pay-phone, pay-station",
+    "708": "pedestal, plinth, footstall",
+    "709": "pencil box, pencil case",
+    "710": "pencil sharpener",
+    "711": "perfume, essence",
+    "712": "Petri dish",
+    "713": "photocopier",
+    "714": "pick, plectrum, plectron",
+    "715": "pickelhaube",
+    "716": "picket fence, paling",
+    "717": "pickup, pickup truck",
+    "718": "pier",
+    "719": "piggy bank, penny bank",
+    "720": "pill bottle",
+    "721": "pillow",
+    "722": "ping-pong ball",
+    "723": "pinwheel",
+    "724": "pirate, pirate ship",
+    "725": "pitcher, ewer",
+    "726": "plane, carpenters plane, woodworking plane",
+    "727": "planetarium",
+    "728": "plastic bag",
+    "729": "plate rack",
+    "730": "plow, plough",
+    "731": "plunger, plumbers helper",
+    "732": "Polaroid camera, Polaroid Land camera",
+    "733": "pole",
+    "734": "police van, police wagon, paddy wagon, patrol wagon, wagon, black Maria",
+    "735": "poncho",
+    "736": "pool table, billiard table, snooker table",
+    "737": "pop bottle, soda bottle",
+    "738": "pot, flowerpot",
+    "739": "potters wheel",
+    "740": "power drill",
+    "741": "prayer rug, prayer mat",
+    "742": "printer",
+    "743": "prison, prison house",
+    "744": "projectile, missile",
+    "745": "projector",
+    "746": "puck, hockey puck",
+    "747": "punching bag, punch bag, punching ball, punchball",
+    "748": "purse",
+    "749": "quill, quill pen",
+    "750": "quilt, comforter, comfort, puff",
+    "751": "racer, race car, racing car",
+    "752": "racket, racquet",
+    "753": "radiator",
+    "754": "radio, wireless",
+    "755": "radio telescope, radio reflector",
+    "756": "rain barrel",
+    "757": "recreational vehicle, RV, R.V.",
+    "758": "reel",
+    "759": "reflex camera",
+    "760": "refrigerator, icebox",
+    "761": "remote control, remote",
+    "762": "restaurant, eating house, eating place, eatery",
+    "763": "revolver, six-gun, six-shooter",
+    "764": "rifle",
+    "765": "rocking chair, rocker",
+    "766": "rotisserie",
+    "767": "rubber eraser, rubber, pencil eraser",
+    "768": "rugby ball",
+    "769": "rule, ruler",
+    "770": "running shoe",
+    "771": "safe",
+    "772": "safety pin",
+    "773": "saltshaker, salt shaker",
+    "774": "sandal",
+    "775": "sarong",
+    "776": "sax, saxophone",
+    "777": "scabbard",
+    "778": "scale, weighing machine",
+    "779": "school bus",
+    "780": "schooner",
+    "781": "scoreboard",
+    "782": "screen, CRT screen",
+    "783": "screw",
+    "784": "screwdriver",
+    "785": "seat belt, seatbelt",
+    "786": "sewing machine",
+    "787": "shield, buckler",
+    "788": "shoe shop, shoe-shop, shoe store",
+    "789": "shoji",
+    "790": "shopping basket",
+    "791": "shopping cart",
+    "792": "shovel",
+    "793": "shower cap",
+    "794": "shower curtain",
+    "795": "ski",
+    "796": "ski mask",
+    "797": "sleeping bag",
+    "798": "slide rule, slipstick",
+    "799": "sliding door",
+    "800": "slot, one-armed bandit",
+    "801": "snorkel",
+    "802": "snowmobile",
+    "803": "snowplow, snowplough",
+    "804": "soap dispenser",
+    "805": "soccer ball",
+    "806": "sock",
+    "807": "solar dish, solar collector, solar furnace",
+    "808": "sombrero",
+    "809": "soup bowl",
+    "810": "space bar",
+    "811": "space heater",
+    "812": "space shuttle",
+    "813": "spatula",
+    "814": "speedboat",
+    "815": "spider web, spiders web",
+    "816": "spindle",
+    "817": "sports car, sport car",
+    "818": "spotlight, spot",
+    "819": "stage",
+    "820": "steam locomotive",
+    "821": "steel arch bridge",
+    "822": "steel drum",
+    "823": "stethoscope",
+    "824": "stole",
+    "825": "stone wall",
+    "826": "stopwatch, stop watch",
+    "827": "stove",
+    "828": "strainer",
+    "829": "streetcar, tram, tramcar, trolley, trolley car",
+    "830": "stretcher",
+    "831": "studio couch, day bed",
+    "832": "stupa, tope",
+    "833": "submarine, pigboat, sub, U-boat",
+    "834": "suit, suit of clothes",
+    "835": "sundial",
+    "836": "sunglass",
+    "837": "sunglasses, dark glasses, shades",
+    "838": "sunscreen, sunblock, sun blocker",
+    "839": "suspension bridge",
+    "840": "swab, swob, mop",
+    "841": "sweatshirt",
+    "842": "swimming trunks, bathing trunks",
+    "843": "swing",
+    "844": "switch, electric switch, electrical switch",
+    "845": "syringe",
+    "846": "table lamp",
+    "847": "tank, army tank, armored combat vehicle, armoured combat vehicle",
+    "848": "tape player",
+    "849": "teapot",
+    "850": "teddy, teddy bear",
+    "851": "television, television system",
+    "852": "tennis ball",
+    "853": "thatch, thatched roof",
+    "854": "theater curtain, theatre curtain",
+    "855": "thimble",
+    "856": "thresher, thrasher, threshing machine",
+    "857": "throne",
+    "858": "tile roof",
+    "859": "toaster",
+    "860": "tobacco shop, tobacconist shop, tobacconist",
+    "861": "toilet seat",
+    "862": "torch",
+    "863": "totem pole",
+    "864": "tow truck, tow car, wrecker",
+    "865": "toyshop",
+    "866": "tractor",
+    "867": "trailer truck, tractor trailer, trucking rig, rig, articulated lorry, semi",
+    "868": "tray",
+    "869": "trench coat",
+    "870": "tricycle, trike, velocipede",
+    "871": "trimaran",
+    "872": "tripod",
+    "873": "triumphal arch",
+    "874": "trolleybus, trolley coach, trackless trolley",
+    "875": "trombone",
+    "876": "tub, vat",
+    "877": "turnstile",
+    "878": "typewriter keyboard",
+    "879": "umbrella",
+    "880": "unicycle, monocycle",
+    "881": "upright, upright piano",
+    "882": "vacuum, vacuum cleaner",
+    "883": "vase",
+    "884": "vault",
+    "885": "velvet",
+    "886": "vending machine",
+    "887": "vestment",
+    "888": "viaduct",
+    "889": "violin, fiddle",
+    "890": "volleyball",
+    "891": "waffle iron",
+    "892": "wall clock",
+    "893": "wallet, billfold, notecase, pocketbook",
+    "894": "wardrobe, closet, press",
+    "895": "warplane, military plane",
+    "896": "washbasin, handbasin, washbowl, lavabo, wash-hand basin",
+    "897": "washer, automatic washer, washing machine",
+    "898": "water bottle",
+    "899": "water jug",
+    "900": "water tower",
+    "901": "whiskey jug",
+    "902": "whistle",
+    "903": "wig",
+    "904": "window screen",
+    "905": "window shade",
+    "906": "Windsor tie",
+    "907": "wine bottle",
+    "908": "wing",
+    "909": "wok",
+    "910": "wooden spoon",
+    "911": "wool, woolen, woollen",
+    "912": "worm fence, snake fence, snake-rail fence, Virginia fence",
+    "913": "wreck",
+    "914": "yawl",
+    "915": "yurt",
+    "916": "web site, website, internet site, site",
+    "917": "comic book",
+    "918": "crossword puzzle, crossword",
+    "919": "street sign",
+    "920": "traffic light, traffic signal, stoplight",
+    "921": "book jacket, dust cover, dust jacket, dust wrapper",
+    "922": "menu",
+    "923": "plate",
+    "924": "guacamole",
+    "925": "consomme",
+    "926": "hot pot, hotpot",
+    "927": "trifle",
+    "928": "ice cream, icecream",
+    "929": "ice lolly, lolly, lollipop, popsicle",
+    "930": "French loaf",
+    "931": "bagel, beigel",
+    "932": "pretzel",
+    "933": "cheeseburger",
+    "934": "hotdog, hot dog, red hot",
+    "935": "mashed potato",
+    "936": "head cabbage",
+    "937": "broccoli",
+    "938": "cauliflower",
+    "939": "zucchini, courgette",
+    "940": "spaghetti squash",
+    "941": "acorn squash",
+    "942": "butternut squash",
+    "943": "cucumber, cuke",
+    "944": "artichoke, globe artichoke",
+    "945": "bell pepper",
+    "946": "cardoon",
+    "947": "mushroom",
+    "948": "Granny Smith",
+    "949": "strawberry",
+    "950": "orange",
+    "951": "lemon",
+    "952": "fig",
+    "953": "pineapple, ananas",
+    "954": "banana",
+    "955": "jackfruit, jak, jack",
+    "956": "custard apple",
+    "957": "pomegranate",
+    "958": "hay",
+    "959": "carbonara",
+    "960": "chocolate sauce, chocolate syrup",
+    "961": "dough",
+    "962": "meat loaf, meatloaf",
+    "963": "pizza, pizza pie",
+    "964": "potpie",
+    "965": "burrito",
+    "966": "red wine",
+    "967": "espresso",
+    "968": "cup",
+    "969": "eggnog",
+    "970": "alp",
+    "971": "bubble",
+    "972": "cliff, drop, drop-off",
+    "973": "coral reef",
+    "974": "geyser",
+    "975": "lakeside, lakeshore",
+    "976": "promontory, headland, head, foreland",
+    "977": "sandbar, sand bar",
+    "978": "seashore, coast, seacoast, sea-coast",
+    "979": "valley, vale",
+    "980": "volcano",
+    "981": "ballplayer, baseball player",
+    "982": "groom, bridegroom",
+    "983": "scuba diver",
+    "984": "rapeseed",
+    "985": "daisy",
+    "986": "yellow ladys slipper, yellow lady-slipper, Cypripedium calceolus, Cypripedium parviflorum",
+    "987": "corn",
+    "988": "acorn",
+    "989": "hip, rose hip, rosehip",
+    "990": "buckeye, horse chestnut, conker",
+    "991": "coral fungus",
+    "992": "agaric",
+    "993": "gyromitra",
+    "994": "stinkhorn, carrion fungus",
+    "995": "earthstar",
+    "996": "hen-of-the-woods, hen of the woods, Polyporus frondosus, Grifola frondosa",
+    "997": "bolete",
+    "998": "ear, spike, capitulum",
+    "999": "toilet tissue, toilet paper, bathroom tissue"
+  }
+}

iMF-B-2/pipeline.py

@@ -0,0 +1,275 @@
+# Copyright 2026 The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+"""Hub custom pipeline: IMFPipeline.
+
+Load with native Hugging Face diffusers and trust_remote_code=True.
+"""
+
+from __future__ import annotations
+
+import json
+from pathlib import Path
+from typing import Dict, List, Optional, Tuple, Union
+
+import torch
+from diffusers.image_processor import VaeImageProcessor
+from diffusers.pipelines.pipeline_utils import DiffusionPipeline, ImagePipelineOutput
+from diffusers.schedulers import FlowMatchEulerDiscreteScheduler
+from diffusers.utils.torch_utils import randn_tensor
+
+
+def _set_imf_timesteps(
+    scheduler: FlowMatchEulerDiscreteScheduler,
+    num_inference_steps: int,
+    device: torch.device,
+) -> torch.Tensor:
+    flow_sigmas = torch.linspace(1.0, 0.0, num_inference_steps + 1, device=device, dtype=torch.float32)
+    scheduler.set_timesteps(sigmas=flow_sigmas.tolist(), device=device)
+    return flow_sigmas
+
+
+class IMFPipeline(DiffusionPipeline):
+    r"""
+    Pipeline for ImageNet class-conditional generation with Improved Mean Flows (iMF).
+
+    Parameters:
+        transformer ([`IMFTransformer2DModel`]):
+            Class-conditioned iMF transformer that predicts mean-flow velocity.
+        scheduler ([`FlowMatchEulerDiscreteScheduler`]):
+            Flow-matching Euler scheduler.
+        vae ([`AutoencoderKL`]):
+            Variational autoencoder used to decode transformer latents to pixels.
+        id2label (`dict[int, str]`, *optional*):
+            ImageNet class id to English label mapping.
+    """
+
+    model_cpu_offload_seq = "transformer->vae"
+
+    def __init__(
+        self,
+        transformer,
+        scheduler,
+        vae,
+        id2label: Optional[Dict[Union[int, str], str]] = None,
+    ):
+        super().__init__()
+        if scheduler is None:
+            scheduler = FlowMatchEulerDiscreteScheduler(
+                num_train_timesteps=1000,
+                shift=1.0,
+                stochastic_sampling=False,
+            )
+        self.register_modules(transformer=transformer, scheduler=scheduler, vae=vae)
+        self.vae_scale_factor = 2 ** (len(self.vae.config.block_out_channels) - 1)
+        self.image_processor = VaeImageProcessor(vae_scale_factor=self.vae_scale_factor)
+        self._id2label = self._normalize_id2label(id2label)
+        self.labels = self._build_label2id(self._id2label)
+        self._labels_loaded_from_model_index = bool(self._id2label)
+
+    def _ensure_labels_loaded(self) -> None:
+        if self._labels_loaded_from_model_index:
+            return
+        loaded = self._read_id2label_from_model_index(getattr(self.config, "_name_or_path", None))
+        if loaded:
+            self._id2label = loaded
+            self.labels = self._build_label2id(self._id2label)
+        self._labels_loaded_from_model_index = True
+
+    @staticmethod
+    def _normalize_id2label(id2label: Optional[Dict[Union[int, str], str]]) -> Dict[int, str]:
+        if not id2label:
+            return {}
+        return {int(key): value for key, value in id2label.items()}
+
+    @staticmethod
+    def _read_id2label_from_model_index(variant_path: Optional[str]) -> Dict[int, str]:
+        if not variant_path:
+            return {}
+        model_index_path = Path(variant_path).resolve() / "model_index.json"
+        if not model_index_path.exists():
+            return {}
+        raw = json.loads(model_index_path.read_text(encoding="utf-8"))
+        id2label = raw.get("id2label")
+        if not isinstance(id2label, dict):
+            return {}
+        return {int(key): value for key, value in id2label.items()}
+
+    @staticmethod
+    def _build_label2id(id2label: Dict[int, str]) -> Dict[str, int]:
+        label2id: Dict[str, int] = {}
+        for class_id, value in id2label.items():
+            for synonym in value.split(","):
+                synonym = synonym.strip()
+                if synonym:
+                    label2id[synonym] = int(class_id)
+        return dict(sorted(label2id.items()))
+
+    @property
+    def id2label(self) -> Dict[int, str]:
+        self._ensure_labels_loaded()
+        return self._id2label
+
+    def get_label_ids(self, label: Union[str, List[str]]) -> List[int]:
+        self._ensure_labels_loaded()
+        if not self.labels:
+            raise ValueError("No labels loaded. Ensure `id2label` exists in model_index.json.")
+        labels = [label] if isinstance(label, str) else label
+        missing = [item for item in labels if item not in self.labels]
+        if missing:
+            preview = ", ".join(list(self.labels.keys())[:8])
+            raise ValueError(f"Unknown label(s): {missing}. Example valid labels: {preview}, ...")
+        return [self.labels[item] for item in labels]
+
+    def _normalize_class_labels(self, class_labels: Union[int, str, List[Union[int, str]]]) -> List[int]:
+        if isinstance(class_labels, int):
+            return [class_labels]
+        if isinstance(class_labels, str):
+            return self.get_label_ids(class_labels)
+        if class_labels and isinstance(class_labels[0], str):
+            return self.get_label_ids(class_labels)
+        return list(class_labels)
+
+    def _predict_velocity_u(
+        self,
+        latents: torch.Tensor,
+        timestep: torch.Tensor,
+        time_gap: torch.Tensor,
+        class_labels: torch.Tensor,
+        class_null: torch.Tensor,
+        guidance_scale: float,
+        guidance_interval_start: float,
+        guidance_interval_end: float,
+        do_classifier_free_guidance: bool,
+    ) -> torch.Tensor:
+        dtype = latents.dtype
+        timestep = timestep.to(device=latents.device, dtype=dtype)
+        time_gap = time_gap.to(device=latents.device, dtype=dtype)
+
+        if do_classifier_free_guidance:
+            latents_in = torch.cat([latents, latents], dim=0)
+            labels = torch.cat([class_labels, class_null], dim=0)
+            omega = torch.tensor([guidance_scale, 1.0], device=latents.device, dtype=dtype)
+            t_min = torch.tensor([guidance_interval_start, 0.0], device=latents.device, dtype=dtype)
+            t_max = torch.tensor([guidance_interval_end, 1.0], device=latents.device, dtype=dtype)
+            batch = latents.shape[0]
+            timestep_in = timestep.reshape(1).repeat(2 * batch)
+            time_gap_in = time_gap.reshape(1).repeat(2 * batch)
+            omega = omega.repeat(batch)
+            t_min = t_min.repeat(batch)
+            t_max = t_max.repeat(batch)
+        else:
+            latents_in = latents
+            labels = class_labels
+            batch = latents.shape[0]
+            timestep_in = timestep.reshape(1).repeat(batch)
+            time_gap_in = time_gap.reshape(1).repeat(batch)
+            omega = torch.full((batch,), guidance_scale, device=latents.device, dtype=dtype)
+            t_min = torch.full((batch,), guidance_interval_start, device=latents.device, dtype=dtype)
+            t_max = torch.full((batch,), guidance_interval_end, device=latents.device, dtype=dtype)
+
+        outputs = self.transformer(
+            sample=latents_in,
+            timestep=timestep_in,
+            class_labels=labels,
+            time_gap=time_gap_in,
+            guidance_scale=omega,
+            guidance_interval_start=t_min,
+            guidance_interval_end=t_max,
+            return_dict=True,
+        )
+        velocity_u = outputs.velocity_u
+
+        if not do_classifier_free_guidance:
+            return velocity_u
+
+        u_cond, u_uncond = velocity_u.chunk(2, dim=0)
+        return u_uncond + guidance_scale * (u_cond - u_uncond)
+
+    def decode_latents(self, latents: torch.Tensor, output_type: str = "pil"):
+        if output_type == "latent":
+            return latents
+
+        scaling_factor = self.vae.config.scaling_factor
+        latents = latents.to(device=self.vae.device, dtype=self.vae.dtype)
+        image = self.vae.decode(latents / scaling_factor).sample
+        if output_type == "pt":
+            return image
+        return self.image_processor.postprocess(image, output_type=output_type)
+
+    @torch.inference_mode()
+    def __call__(
+        self,
+        class_labels: Union[int, str, List[Union[int, str]]],
+        num_inference_steps: int = 1,
+        guidance_scale: float = 2.7,
+        guidance_interval_start: float = 0.1,
+        guidance_interval_end: float = 0.9,
+        generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
+        latents: Optional[torch.Tensor] = None,
+        output_type: str = "pil",
+        return_dict: bool = True,
+    ) -> Union[ImagePipelineOutput, Tuple]:
+        if output_type not in {"pil", "np", "pt", "latent"}:
+            raise ValueError("output_type must be one of: 'pil', 'np', 'pt', 'latent'.")
+
+        class_label_ids = self._normalize_class_labels(class_labels)
+        do_classifier_free_guidance = guidance_scale > 1.0
+        batch_size = len(class_label_ids)
+
+        image_size = int(self.transformer.config.sample_size)
+        channels = int(self.transformer.config.in_channels)
+        null_class_val = int(self.transformer.config.num_classes)
+
+        if latents is None:
+            latents = randn_tensor(
+                shape=(batch_size, channels, image_size, image_size),
+                generator=generator,
+                device=self._execution_device,
+                dtype=self.transformer.dtype,
+            )
+
+        class_labels_t = torch.tensor(class_label_ids, device=latents.device, dtype=torch.long).reshape(-1)
+        class_labels_t = class_labels_t.clamp(0, null_class_val - 1)
+        class_null = torch.full_like(class_labels_t, null_class_val)
+
+        flow_sigmas = _set_imf_timesteps(self.scheduler, num_inference_steps, latents.device)
+
+        for i in self.progress_bar(range(num_inference_steps)):
+            t = flow_sigmas[i]
+            t_next = flow_sigmas[i + 1]
+            time_gap = t - t_next
+            velocity_u = self._predict_velocity_u(
+                latents,
+                t,
+                time_gap,
+                class_labels_t,
+                class_null,
+                guidance_scale,
+                guidance_interval_start,
+                guidance_interval_end,
+                do_classifier_free_guidance,
+            )
+            latents = self.scheduler.step(velocity_u, self.scheduler.timesteps[i], latents).prev_sample
+
+        images = self.decode_latents(latents, output_type=output_type)
+
+        self.maybe_free_model_hooks()
+
+        if not return_dict:
+            return (images,)
+        return ImagePipelineOutput(images=images)
+
+
+IMFPipelineOutput = ImagePipelineOutput

iMF-B-2/scheduler/scheduler_config.json

@@ -0,0 +1,7 @@
+{
+  "_class_name": "FlowMatchEulerDiscreteScheduler",
+  "_diffusers_version": "0.36.0",
+  "num_train_timesteps": 1000,
+  "shift": 1.0,
+  "stochastic_sampling": false
+}

iMF-B-2/transformer/config.json

@@ -0,0 +1,23 @@
+{
+  "_class_name": "IMFTransformer2DModel",
+  "_diffusers_version": "0.38.0",
+  "aux_head_depth": 8,
+  "depth": 12,
+  "embedding_init_constant": 1.0,
+  "eval_mode": true,
+  "hidden_size": 768,
+  "in_channels": 4,
+  "mlp_ratio": 2.6666666666666665,
+  "model_type": "iMF-B/2",
+  "num_attention_heads": 12,
+  "num_cfg_tokens": 4,
+  "num_class_embeds": null,
+  "num_class_tokens": 8,
+  "num_classes": 1000,
+  "num_interval_tokens": 2,
+  "num_time_tokens": 4,
+  "patch_size": 2,
+  "sample_size": 32,
+  "token_init_constant": 1.0,
+  "weight_init_constant": 0.32
+}

iMF-B-2/transformer/diffusion_pytorch_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a3f6ec4ba47a9acbc2e183482750f3c7ed6856ea0ee258cf44edfbf39fbe64cc
+size 355811968

iMF-B-2/transformer/transformer_imf.py

@@ -0,0 +1,586 @@
+from __future__ import annotations
+
+import argparse
+import math
+from collections.abc import Mapping
+from dataclasses import dataclass
+from functools import partial
+from math import sqrt
+from typing import Dict, Literal, Optional, Tuple, Union
+
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+from diffusers.configuration_utils import ConfigMixin, register_to_config
+from diffusers.models.modeling_utils import ModelMixin
+from diffusers.utils import BaseOutput
+
+
+IMF_PRESET_CONFIGS: Dict[str, Dict[str, object]] = {
+    "iMF-B/2": {
+        "sample_size": 32,
+        "patch_size": 2,
+        "hidden_size": 768,
+        "depth": 12,
+        "num_attention_heads": 12,
+        "aux_head_depth": 8,
+    },
+    "iMF-M/2": {
+        "sample_size": 32,
+        "patch_size": 2,
+        "hidden_size": 768,
+        "depth": 24,
+        "num_attention_heads": 12,
+        "aux_head_depth": 8,
+    },
+    "iMF-L/2": {
+        "sample_size": 32,
+        "patch_size": 2,
+        "hidden_size": 1024,
+        "depth": 32,
+        "num_attention_heads": 16,
+        "aux_head_depth": 8,
+    },
+    "iMF-XL/2": {
+        "sample_size": 32,
+        "patch_size": 2,
+        "hidden_size": 1024,
+        "depth": 48,
+        "num_attention_heads": 16,
+        "aux_head_depth": 8,
+    },
+}
+
+
+@dataclass
+class IMFTransformer2DModelOutput(BaseOutput):
+    velocity_u: torch.Tensor
+    velocity_v: Optional[torch.Tensor] = None
+
+
+def remap_legacy_state_dict(state_dict: Dict[str, torch.Tensor]) -> Dict[str, torch.Tensor]:
+    """Map legacy iMF / Flax-style keys to native IMFTransformer2DModel keys."""
+    remapped: Dict[str, torch.Tensor] = {}
+    for key, value in state_dict.items():
+        new_key = key
+        for prefix in ("net.", "transformer."):
+            if new_key.startswith(prefix):
+                new_key = new_key[len(prefix) :]
+                break
+        new_key = new_key.replace(".kernel", ".weight")
+        new_key = new_key.replace("._flax_linear.", ".linear.")
+        new_key = new_key.replace("._flax_embedding.", ".embedding.")
+        remapped[new_key] = value
+    return remapped
+
+
+def config_from_legacy(config: Dict[str, object]) -> Dict[str, object]:
+    model_type = config.get("model_type") or config.get("model_str") or config.get("model_name")
+    if model_type not in IMF_PRESET_CONFIGS:
+        raise ValueError(f"Unknown iMF preset '{model_type}'. Known: {list(IMF_PRESET_CONFIGS)}")
+
+    preset = dict(IMF_PRESET_CONFIGS[model_type])
+    preset["num_classes"] = int(config.get("num_class_embeds") or config.get("num_classes") or 1000)
+    preset["model_type"] = model_type
+    if config.get("sample_size") is not None:
+        preset["sample_size"] = int(config["sample_size"])
+    if config.get("in_channels") is not None:
+        preset["in_channels"] = int(config["in_channels"])
+    return preset
+
+
+def _unsqueeze(t: torch.Tensor, dim: int) -> torch.Tensor:
+    return t.unsqueeze(dim)
+
+
+class _ScaledLinear(nn.Module):
+    def __init__(
+        self,
+        in_features: int,
+        out_features: int,
+        bias: bool = True,
+        weight_init: str = "scaled_variance",
+        init_constant: float = 1.0,
+        bias_init: str = "zeros",
+    ):
+        super().__init__()
+        self.linear = nn.Linear(in_features, out_features, bias=bias)
+        if weight_init == "scaled_variance":
+            std = init_constant / sqrt(in_features)
+            nn.init.normal_(self.linear.weight, std=std)
+        elif weight_init == "zeros":
+            nn.init.zeros_(self.linear.weight)
+        else:
+            raise ValueError(f"Invalid weight_init: {weight_init}")
+        if bias:
+            if bias_init == "zeros":
+                nn.init.zeros_(self.linear.bias)
+            else:
+                raise ValueError(f"Invalid bias_init: {bias_init}")
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        return self.linear(x)
+
+
+class _ScaledEmbedding(nn.Module):
+    def __init__(
+        self,
+        num_embeddings: int,
+        embedding_dim: int,
+        weight_init: str = "scaled_variance",
+        init_constant: float = 1.0,
+    ):
+        super().__init__()
+        self.embedding = nn.Embedding(num_embeddings, embedding_dim)
+        if weight_init == "scaled_variance":
+            std = init_constant / sqrt(embedding_dim)
+            nn.init.normal_(self.embedding.weight, std=std)
+        else:
+            raise ValueError(f"Invalid weight_init: {weight_init}")
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        return self.embedding(x)
+
+
+class _RMSNorm(nn.Module):
+    def __init__(self, dim: int, eps: float = 1e-6):
+        super().__init__()
+        self.weight = nn.Parameter(torch.ones(dim))
+        self.eps = eps
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        mean_square = torch.mean(torch.square(x), dim=-1, keepdim=True)
+        output = x * torch.rsqrt(mean_square + self.eps)
+        return output.to(x.dtype) * self.weight
+
+
+class _SwiGLUMlp(nn.Module):
+    def __init__(
+        self,
+        in_features: int,
+        hidden_features: int,
+        weight_init: str = "scaled_variance",
+        weight_init_constant: float = 1.0,
+    ):
+        super().__init__()
+        init_kwargs = dict(bias=False, weight_init=weight_init, init_constant=weight_init_constant)
+        self.w1 = _ScaledLinear(in_features, hidden_features, **init_kwargs)
+        self.w3 = _ScaledLinear(in_features, hidden_features, **init_kwargs)
+        self.w2 = _ScaledLinear(hidden_features, in_features, **init_kwargs)
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        return self.w2(F.silu(self.w1(x)) * self.w3(x))
+
+
+class IMFTimestepEmbedder(nn.Module):
+    def __init__(
+        self,
+        hidden_size: int,
+        frequency_embedding_size: int = 256,
+        init_constant: float = 1.0,
+    ):
+        super().__init__()
+        self.mlp = nn.Sequential(
+            _ScaledLinear(frequency_embedding_size, hidden_size, bias=True, init_constant=init_constant),
+            nn.SiLU(),
+            _ScaledLinear(hidden_size, hidden_size, bias=True, init_constant=init_constant),
+        )
+        self.frequency_embedding_size = frequency_embedding_size
+
+    @staticmethod
+    def timestep_embedding(t: torch.Tensor, dim: int, max_period: int = 10000) -> torch.Tensor:
+        half = dim // 2
+        freqs = torch.exp(
+            -math.log(max_period) * torch.arange(start=0, end=half, dtype=torch.float32, device=t.device) / half
+        )
+        args = t[:, None].float() * freqs[None]
+        embedding = torch.cat([torch.cos(args), torch.sin(args)], dim=-1)
+        if dim % 2:
+            embedding = torch.cat([embedding, torch.zeros_like(embedding[:, :1])], dim=-1)
+        return embedding
+
+    def forward(self, t: torch.Tensor) -> torch.Tensor:
+        embedding = self.timestep_embedding(t, self.frequency_embedding_size)
+        return self.mlp(embedding.to(dtype=next(self.mlp.parameters()).dtype))
+
+
+class IMFLabelEmbedder(nn.Module):
+    def __init__(self, num_classes: int, hidden_size: int, init_constant: float = 1.0):
+        super().__init__()
+        self.embedding_table = _ScaledEmbedding(num_classes + 1, hidden_size, init_constant=init_constant)
+
+    def forward(self, labels: torch.Tensor) -> torch.Tensor:
+        return self.embedding_table(labels)
+
+
+class IMFPatchEmbedder(nn.Module):
+    def __init__(self, input_size: int, patch_size: int, in_channels: int, hidden_size: int, bias: bool = True):
+        super().__init__()
+        self.patch_size = (patch_size, patch_size)
+        self.num_patches = (input_size // patch_size) ** 2
+        self.proj = nn.Conv2d(
+            in_channels,
+            hidden_size,
+            kernel_size=patch_size,
+            stride=patch_size,
+            bias=bias,
+        )
+        kh = kw = patch_size
+        fan_in = kh * kw * in_channels
+        fan_out = hidden_size
+        limit = math.sqrt(6.0 / (fan_in + fan_out))
+        nn.init.uniform_(self.proj.weight, -limit, limit)
+        if bias:
+            nn.init.zeros_(self.proj.bias)
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        x = self.proj(x)
+        return x.flatten(2).transpose(1, 2)
+
+
+def precompute_rope_freqs(dim: int, seq_len: int, theta: float = 10000.0, device: torch.device | None = None):
+    freqs = 1.0 / (theta ** (torch.arange(0, dim, 2, dtype=torch.float32, device=device) / dim))
+    positions = torch.arange(seq_len, dtype=torch.float32, device=device)
+    freqs_cis = torch.outer(positions, freqs)
+    return torch.complex(torch.cos(freqs_cis), torch.sin(freqs_cis))
+
+
+def apply_rotary_pos_emb(x: torch.Tensor, freqs_cis: torch.Tensor) -> torch.Tensor:
+    x_ = x.float().reshape(*x.shape[:-1], -1, 2)
+    x_complex = torch.view_as_complex(x_)
+    freqs_cis = freqs_cis.unsqueeze(0).unsqueeze(2).to(x.device)
+    x_rotated = torch.view_as_real(x_complex * freqs_cis).flatten(-2)
+    return x_rotated.to(dtype=x.dtype)
+
+
+class IMFRoPEAttention(nn.Module):
+    def __init__(self, hidden_size: int, num_heads: int, weight_init_constant: float = 0.32):
+        super().__init__()
+        self.hidden_size = hidden_size
+        self.num_heads = num_heads
+        self.head_dim = hidden_size // num_heads
+        init_kwargs = dict(bias=False, init_constant=weight_init_constant)
+        self.q_proj = _ScaledLinear(hidden_size, hidden_size, **init_kwargs)
+        self.k_proj = _ScaledLinear(hidden_size, hidden_size, **init_kwargs)
+        self.v_proj = _ScaledLinear(hidden_size, hidden_size, **init_kwargs)
+        self.out_proj = _ScaledLinear(hidden_size, hidden_size, **init_kwargs)
+        self.q_norm = _RMSNorm(self.head_dim)
+        self.k_norm = _RMSNorm(self.head_dim)
+
+    def forward(self, x: torch.Tensor, rope_freqs: torch.Tensor) -> torch.Tensor:
+        batch, seq_len, _ = x.shape
+        q = self.q_proj(x).reshape(batch, seq_len, self.num_heads, self.head_dim)
+        k = self.k_proj(x).reshape(batch, seq_len, self.num_heads, self.head_dim)
+        v = self.v_proj(x).reshape(batch, seq_len, self.num_heads, self.head_dim)
+        q = apply_rotary_pos_emb(self.q_norm(q), rope_freqs)
+        k = apply_rotary_pos_emb(self.k_norm(k), rope_freqs)
+        query = q / math.sqrt(self.head_dim)
+        attn_weights = torch.einsum("bqhd,bkhd->bhqk", query, k)
+        attn_weights = F.softmax(attn_weights, dim=-1, dtype=torch.float32).to(v.dtype)
+        attn = torch.einsum("bhqk,bkhd->bqhd", attn_weights, v)
+        return self.out_proj(attn.reshape(batch, seq_len, self.hidden_size))
+
+
+class IMFTransformerBlock(nn.Module):
+    def __init__(self, hidden_size: int, num_heads: int, mlp_ratio: float, weight_init_constant: float):
+        super().__init__()
+        self.norm1 = _RMSNorm(hidden_size)
+        self.attn = IMFRoPEAttention(hidden_size, num_heads, weight_init_constant=weight_init_constant)
+        self.norm2 = _RMSNorm(hidden_size)
+        mlp_hidden_dim = int(hidden_size * mlp_ratio)
+        self.mlp = _SwiGLUMlp(hidden_size, mlp_hidden_dim, weight_init_constant=weight_init_constant)
+        self.attn_scale = nn.Parameter(torch.zeros(hidden_size))
+        self.mlp_scale = nn.Parameter(torch.zeros(hidden_size))
+
+    def forward(self, x: torch.Tensor, rope_freqs: torch.Tensor) -> torch.Tensor:
+        x = x + self.attn(self.norm1(x), rope_freqs) * self.attn_scale
+        x = x + self.mlp(self.norm2(x)) * self.mlp_scale
+        return x
+
+
+class IMFFinalLayer(nn.Module):
+    def __init__(self, hidden_size: int, patch_size: int, out_channels: int):
+        super().__init__()
+        self.norm = _RMSNorm(hidden_size)
+        self.linear = _ScaledLinear(
+            hidden_size,
+            patch_size * patch_size * out_channels,
+            bias=True,
+            weight_init="zeros",
+            init_constant=1.0,
+        )
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        return self.linear(self.norm(x))
+
+
+class IMFTransformer2DModel(ModelMixin, ConfigMixin):
+    _supports_gradient_checkpointing = True
+
+    @register_to_config
+    def __init__(
+        self,
+        sample_size: int = 32,
+        patch_size: int = 2,
+        in_channels: int = 4,
+        hidden_size: int = 768,
+        depth: int = 12,
+        num_attention_heads: int = 12,
+        mlp_ratio: float = 8 / 3,
+        num_classes: int = 1000,
+        aux_head_depth: int = 8,
+        num_class_tokens: int = 8,
+        num_time_tokens: int = 4,
+        num_cfg_tokens: int = 4,
+        num_interval_tokens: int = 2,
+        token_init_constant: float = 1.0,
+        embedding_init_constant: float = 1.0,
+        weight_init_constant: float = 0.32,
+        eval_mode: bool = True,
+        model_type: str | None = None,
+        num_class_embeds: int | None = None,
+    ):
+        super().__init__()
+        if num_class_embeds is not None:
+            num_classes = int(num_class_embeds)
+        if model_type in IMF_PRESET_CONFIGS:
+            preset = IMF_PRESET_CONFIGS[model_type]
+            sample_size = int(preset["sample_size"])
+            patch_size = int(preset["patch_size"])
+            hidden_size = int(preset["hidden_size"])
+            depth = int(preset["depth"])
+            num_attention_heads = int(preset["num_attention_heads"])
+            aux_head_depth = int(preset["aux_head_depth"])
+
+        self.sample_size = sample_size
+        self.patch_size = patch_size
+        self.in_channels = in_channels
+        self.out_channels = in_channels
+        self.hidden_size = hidden_size
+        self.depth = depth
+        self.num_attention_heads = num_attention_heads
+        self.num_classes = num_classes
+        self.aux_head_depth = aux_head_depth
+        self.num_class_tokens = num_class_tokens
+        self.num_time_tokens = num_time_tokens
+        self.num_cfg_tokens = num_cfg_tokens
+        self.num_interval_tokens = num_interval_tokens
+        self.eval_mode = eval_mode
+        self.gradient_checkpointing = False
+
+        self.x_embedder = IMFPatchEmbedder(sample_size, patch_size, in_channels, hidden_size, bias=True)
+        embed_kwargs = dict(hidden_size=hidden_size, init_constant=embedding_init_constant)
+        self.h_embedder = IMFTimestepEmbedder(**embed_kwargs)
+        self.omega_embedder = IMFTimestepEmbedder(**embed_kwargs)
+        self.cfg_t_start_embedder = IMFTimestepEmbedder(**embed_kwargs)
+        self.cfg_t_end_embedder = IMFTimestepEmbedder(**embed_kwargs)
+        self.y_embedder = IMFLabelEmbedder(num_classes, hidden_size, init_constant=embedding_init_constant)
+
+        token_std = token_init_constant / math.sqrt(hidden_size)
+        self.time_tokens = nn.Parameter(torch.randn(num_time_tokens, hidden_size) * token_std)
+        self.class_tokens = nn.Parameter(torch.randn(num_class_tokens, hidden_size) * token_std)
+        self.omega_tokens = nn.Parameter(torch.randn(num_cfg_tokens, hidden_size) * token_std)
+        self.t_min_tokens = nn.Parameter(torch.randn(num_interval_tokens, hidden_size) * token_std)
+        self.t_max_tokens = nn.Parameter(torch.randn(num_interval_tokens, hidden_size) * token_std)
+
+        total_tokens = (
+            self.x_embedder.num_patches
+            + num_class_tokens
+            + num_cfg_tokens
+            + 2 * num_interval_tokens
+            + num_time_tokens
+        )
+        self.prefix_tokens = num_class_tokens + num_cfg_tokens + 2 * num_interval_tokens + num_time_tokens
+        self.head_dim = hidden_size // num_attention_heads
+        self.register_buffer(
+            "rope_freqs",
+            precompute_rope_freqs(self.head_dim, total_tokens),
+            persistent=False,
+        )
+
+        shared_depth = depth - aux_head_depth
+        block_kwargs = dict(
+            hidden_size=hidden_size,
+            num_heads=num_attention_heads,
+            mlp_ratio=mlp_ratio,
+            weight_init_constant=weight_init_constant,
+        )
+        self.shared_blocks = nn.ModuleList([IMFTransformerBlock(**block_kwargs) for _ in range(shared_depth)])
+        self.u_heads = nn.ModuleList([IMFTransformerBlock(**block_kwargs) for _ in range(aux_head_depth)])
+        self.v_heads = nn.ModuleList(
+            [IMFTransformerBlock(**block_kwargs) for _ in range(aux_head_depth if not eval_mode else 0)]
+        )
+        self.u_final_layer = IMFFinalLayer(hidden_size, patch_size, in_channels)
+        self.v_final_layer = IMFFinalLayer(hidden_size, patch_size, in_channels)
+
+    def unpatchify(self, x: torch.Tensor) -> torch.Tensor:
+        c = self.out_channels
+        p = self.patch_size
+        h = w = int(x.shape[1] ** 0.5)
+        x = x.reshape(x.shape[0], h, w, p, p, c)
+        x = torch.einsum("nhwpqc->nchpwq", x)
+        return x.reshape(x.shape[0], c, h * p, w * p)
+
+    def _build_sequence(
+        self,
+        sample: torch.Tensor,
+        time_gap: torch.Tensor,
+        guidance_scale: torch.Tensor,
+        guidance_interval_start: torch.Tensor,
+        guidance_interval_end: torch.Tensor,
+        class_labels: torch.Tensor,
+    ) -> torch.Tensor:
+        x_embed = self.x_embedder(sample)
+        h_embed = self.h_embedder(time_gap)
+        omega_embed = self.omega_embedder(1 - 1 / guidance_scale)
+        t_min_embed = self.cfg_t_start_embedder(guidance_interval_start)
+        t_max_embed = self.cfg_t_end_embedder(guidance_interval_end)
+        y_embed = self.y_embedder(class_labels)
+
+        time_tokens = self.time_tokens + _unsqueeze(h_embed, 1)
+        omega_tokens = self.omega_tokens + _unsqueeze(omega_embed, 1)
+        t_min_tokens = self.t_min_tokens + _unsqueeze(t_min_embed, 1)
+        t_max_tokens = self.t_max_tokens + _unsqueeze(t_max_embed, 1)
+        class_tokens = self.class_tokens + _unsqueeze(y_embed, 1)
+
+        return torch.cat(
+            [class_tokens, omega_tokens, t_min_tokens, t_max_tokens, time_tokens, x_embed],
+            dim=1,
+        )
+
+    def forward(
+        self,
+        sample: torch.Tensor,
+        timestep: torch.Tensor,
+        class_labels: torch.Tensor,
+        time_gap: Optional[torch.Tensor] = None,
+        guidance_scale: Optional[torch.Tensor] = None,
+        guidance_interval_start: Optional[torch.Tensor] = None,
+        guidance_interval_end: Optional[torch.Tensor] = None,
+        return_dict: bool = True,
+    ) -> Union[IMFTransformer2DModelOutput, Tuple[torch.Tensor, ...]]:
+        batch_size = sample.shape[0]
+        timestep = self._expand_batch(timestep, batch_size, sample.device, sample.dtype)
+        if time_gap is None:
+            time_gap = timestep
+        else:
+            time_gap = self._expand_batch(time_gap, batch_size, sample.device, sample.dtype)
+        if guidance_scale is None:
+            guidance_scale = torch.ones(batch_size, device=sample.device, dtype=sample.dtype)
+        else:
+            guidance_scale = self._expand_batch(guidance_scale, batch_size, sample.device, sample.dtype)
+        if guidance_interval_start is None:
+            guidance_interval_start = torch.zeros(batch_size, device=sample.device, dtype=sample.dtype)
+        else:
+            guidance_interval_start = self._expand_batch(
+                guidance_interval_start, batch_size, sample.device, sample.dtype
+            )
+        if guidance_interval_end is None:
+            guidance_interval_end = torch.ones(batch_size, device=sample.device, dtype=sample.dtype)
+        else:
+            guidance_interval_end = self._expand_batch(
+                guidance_interval_end, batch_size, sample.device, sample.dtype
+            )
+
+        class_labels = class_labels.reshape(-1).long()
+        seq = self._build_sequence(
+            sample,
+            time_gap,
+            guidance_scale,
+            guidance_interval_start,
+            guidance_interval_end,
+            class_labels,
+        )
+
+        for block in self.shared_blocks:
+            if self.training and self.gradient_checkpointing:
+                seq = torch.utils.checkpoint.checkpoint(block, seq, self.rope_freqs, use_reentrant=False)
+            else:
+                seq = block(seq, self.rope_freqs)
+
+        u_seq = v_seq = seq
+        for block in self.u_heads:
+            u_seq = block(u_seq, self.rope_freqs)
+        for block in self.v_heads:
+            v_seq = block(v_seq, self.rope_freqs)
+
+        u_tokens = u_seq[:, self.prefix_tokens :]
+        velocity_u = self.unpatchify(self.u_final_layer(u_tokens))
+
+        velocity_v = None
+        if len(self.v_heads) > 0:
+            v_tokens = v_seq[:, self.prefix_tokens :]
+            velocity_v = self.unpatchify(self.v_final_layer(v_tokens))
+
+        if not return_dict:
+            if velocity_v is None:
+                return (velocity_u,)
+            return (velocity_u, velocity_v)
+        return IMFTransformer2DModelOutput(velocity_u=velocity_u, velocity_v=velocity_v)
+
+    @staticmethod
+    def _expand_batch(
+        value: torch.Tensor,
+        batch_size: int,
+        device: torch.device,
+        dtype: torch.dtype,
+    ) -> torch.Tensor:
+        value = torch.as_tensor(value, device=device, dtype=dtype)
+        if value.ndim == 0:
+            value = value.reshape(1).repeat(batch_size)
+        else:
+            value = value.reshape(-1)
+            if value.shape[0] == 1 and batch_size > 1:
+                value = value.repeat(batch_size)
+        return value
+
+    @classmethod
+    def from_imf_checkpoint(
+        cls,
+        checkpoint_path: str,
+        model_type: str | None = None,
+        map_location: str = "cpu",
+        strict: bool = True,
+        eval_mode: bool = True,
+    ) -> Tuple["IMFTransformer2DModel", Dict[str, object]]:
+        checkpoint = torch.load(checkpoint_path, map_location=map_location, weights_only=False)
+        if isinstance(checkpoint, Mapping) and "state_dict" in checkpoint:
+            state_dict = checkpoint["state_dict"]
+        elif isinstance(checkpoint, Mapping) and any(k.startswith("net.") for k in checkpoint):
+            state_dict = checkpoint
+        else:
+            state_dict = checkpoint
+
+        inferred_type = model_type
+        if inferred_type is None and isinstance(checkpoint, Mapping):
+            args = checkpoint.get("args")
+            if isinstance(args, argparse.Namespace):
+                inferred_type = getattr(args, "model", None) or getattr(args, "model_str", None)
+            elif isinstance(args, Mapping):
+                inferred_type = args.get("model") or args.get("model_str")
+
+        if inferred_type is None:
+            raise ValueError("model_type must be provided when loading a raw state dict checkpoint.")
+
+        config = dict(IMF_PRESET_CONFIGS[inferred_type])
+        config["model_type"] = inferred_type
+        config["eval_mode"] = eval_mode
+        model = cls(**config)
+        model.load_state_dict(remap_legacy_state_dict(state_dict), strict=strict)
+        metadata = {"checkpoint_path": checkpoint_path, "model_type": inferred_type}
+        return model, metadata
+
+    def to_imf_checkpoint(self, prefix: str = "net.") -> Dict[str, torch.Tensor]:
+        state: Dict[str, torch.Tensor] = {}
+        for key, value in self.state_dict().items():
+            if key == "rope_freqs":
+                continue
+            state[f"{prefix}{key}"] = value.detach().cpu()
+        return state
+
+    @property
+    def net(self):
+        return self
+
+
+IMFDiffusersModel = IMFTransformer2DModel

iMF-B-2/vae/config.json

@@ -0,0 +1,38 @@
+{
+  "_class_name": "AutoencoderKL",
+  "_diffusers_version": "0.36.0",
+  "_name_or_path": "stabilityai/sd-vae-ft-mse",
+  "act_fn": "silu",
+  "block_out_channels": [
+    128,
+    256,
+    512,
+    512
+  ],
+  "down_block_types": [
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D"
+  ],
+  "force_upcast": true,
+  "in_channels": 3,
+  "latent_channels": 4,
+  "latents_mean": null,
+  "latents_std": null,
+  "layers_per_block": 2,
+  "mid_block_add_attention": true,
+  "norm_num_groups": 32,
+  "out_channels": 3,
+  "sample_size": 256,
+  "scaling_factor": 0.18215,
+  "shift_factor": null,
+  "up_block_types": [
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D"
+  ],
+  "use_post_quant_conv": true,
+  "use_quant_conv": true
+}

iMF-B-2/vae/diffusion_pytorch_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a1d993488569e928462932c8c38a0760b874d166399b14414135bd9c42df5815
+size 334643276

iMF-L-2/README.md

@@ -0,0 +1,40 @@
+---
+license: mit
+library_name: diffusers
+pipeline_tag: text-to-image
+tags:
+- diffusers
+- imf
+- image-generation
+- class-conditional
+---
+
+# iMF-L-2
+
+Self-contained Diffusers variant for **iMF-L/2** (Improved Mean Flows).
+
+## Load
+
+```python
+from pathlib import Path
+from diffusers import DiffusionPipeline
+import torch
+
+model_dir = Path("iMF-L-2")
+pipe = DiffusionPipeline.from_pretrained(
+    str(model_dir),
+    local_files_only=True,
+    custom_pipeline=str(model_dir / "pipeline.py"),
+    trust_remote_code=True,
+    torch_dtype=torch.float32,
+)
+pipe.to("cuda")
+
+image = pipe(
+    class_labels=207,
+    num_inference_steps=1,
+    guidance_scale=2.7,
+    guidance_interval_start=0.1,
+    guidance_interval_end=0.9,
+).images[0]
+```

iMF-L-2/conversion_metadata.json

@@ -0,0 +1,8 @@
+{
+  "checkpoint_path": "/data/projects/Visual-Generative-Foundation-Model-Collection/models/BiliSakura/iMF-diffusers/iMF-L.pth",
+  "model_type": "iMF-L/2",
+  "imf_args": {
+    "model_type": "iMF-L/2",
+    "model_str": "iMF-L/2"
+  }
+}

iMF-L-2/model_index.json

@@ -0,0 +1,1021 @@
+{
+  "_class_name": [
+    "pipeline",
+    "IMFPipeline"
+  ],
+  "_diffusers_version": "0.36.0",
+  "scheduler": [
+    "diffusers",
+    "FlowMatchEulerDiscreteScheduler"
+  ],
+  "transformer": [
+    "transformer_imf",
+    "IMFTransformer2DModel"
+  ],
+  "vae": [
+    "diffusers",
+    "AutoencoderKL"
+  ],
+  "id2label": {
+    "0": "tench, Tinca tinca",
+    "1": "goldfish, Carassius auratus",
+    "2": "great white shark, white shark, man-eater, man-eating shark, Carcharodon carcharias",
+    "3": "tiger shark, Galeocerdo cuvieri",
+    "4": "hammerhead, hammerhead shark",
+    "5": "electric ray, crampfish, numbfish, torpedo",
+    "6": "stingray",
+    "7": "cock",
+    "8": "hen",
+    "9": "ostrich, Struthio camelus",
+    "10": "brambling, Fringilla montifringilla",
+    "11": "goldfinch, Carduelis carduelis",
+    "12": "house finch, linnet, Carpodacus mexicanus",
+    "13": "junco, snowbird",
+    "14": "indigo bunting, indigo finch, indigo bird, Passerina cyanea",
+    "15": "robin, American robin, Turdus migratorius",
+    "16": "bulbul",
+    "17": "jay",
+    "18": "magpie",
+    "19": "chickadee",
+    "20": "water ouzel, dipper",
+    "21": "kite",
+    "22": "bald eagle, American eagle, Haliaeetus leucocephalus",
+    "23": "vulture",
+    "24": "great grey owl, great gray owl, Strix nebulosa",
+    "25": "European fire salamander, Salamandra salamandra",
+    "26": "common newt, Triturus vulgaris",
+    "27": "eft",
+    "28": "spotted salamander, Ambystoma maculatum",
+    "29": "axolotl, mud puppy, Ambystoma mexicanum",
+    "30": "bullfrog, Rana catesbeiana",
+    "31": "tree frog, tree-frog",
+    "32": "tailed frog, bell toad, ribbed toad, tailed toad, Ascaphus trui",
+    "33": "loggerhead, loggerhead turtle, Caretta caretta",
+    "34": "leatherback turtle, leatherback, leathery turtle, Dermochelys coriacea",
+    "35": "mud turtle",
+    "36": "terrapin",
+    "37": "box turtle, box tortoise",
+    "38": "banded gecko",
+    "39": "common iguana, iguana, Iguana iguana",
+    "40": "American chameleon, anole, Anolis carolinensis",
+    "41": "whiptail, whiptail lizard",
+    "42": "agama",
+    "43": "frilled lizard, Chlamydosaurus kingi",
+    "44": "alligator lizard",
+    "45": "Gila monster, Heloderma suspectum",
+    "46": "green lizard, Lacerta viridis",
+    "47": "African chameleon, Chamaeleo chamaeleon",
+    "48": "Komodo dragon, Komodo lizard, dragon lizard, giant lizard, Varanus komodoensis",
+    "49": "African crocodile, Nile crocodile, Crocodylus niloticus",
+    "50": "American alligator, Alligator mississipiensis",
+    "51": "triceratops",
+    "52": "thunder snake, worm snake, Carphophis amoenus",
+    "53": "ringneck snake, ring-necked snake, ring snake",
+    "54": "hognose snake, puff adder, sand viper",
+    "55": "green snake, grass snake",
+    "56": "king snake, kingsnake",
+    "57": "garter snake, grass snake",
+    "58": "water snake",
+    "59": "vine snake",
+    "60": "night snake, Hypsiglena torquata",
+    "61": "boa constrictor, Constrictor constrictor",
+    "62": "rock python, rock snake, Python sebae",
+    "63": "Indian cobra, Naja naja",
+    "64": "green mamba",
+    "65": "sea snake",
+    "66": "horned viper, cerastes, sand viper, horned asp, Cerastes cornutus",
+    "67": "diamondback, diamondback rattlesnake, Crotalus adamanteus",
+    "68": "sidewinder, horned rattlesnake, Crotalus cerastes",
+    "69": "trilobite",
+    "70": "harvestman, daddy longlegs, Phalangium opilio",
+    "71": "scorpion",
+    "72": "black and gold garden spider, Argiope aurantia",
+    "73": "barn spider, Araneus cavaticus",
+    "74": "garden spider, Aranea diademata",
+    "75": "black widow, Latrodectus mactans",
+    "76": "tarantula",
+    "77": "wolf spider, hunting spider",
+    "78": "tick",
+    "79": "centipede",
+    "80": "black grouse",
+    "81": "ptarmigan",
+    "82": "ruffed grouse, partridge, Bonasa umbellus",
+    "83": "prairie chicken, prairie grouse, prairie fowl",
+    "84": "peacock",
+    "85": "quail",
+    "86": "partridge",
+    "87": "African grey, African gray, Psittacus erithacus",
+    "88": "macaw",
+    "89": "sulphur-crested cockatoo, Kakatoe galerita, Cacatua galerita",
+    "90": "lorikeet",
+    "91": "coucal",
+    "92": "bee eater",
+    "93": "hornbill",
+    "94": "hummingbird",
+    "95": "jacamar",
+    "96": "toucan",
+    "97": "drake",
+    "98": "red-breasted merganser, Mergus serrator",
+    "99": "goose",
+    "100": "black swan, Cygnus atratus",
+    "101": "tusker",
+    "102": "echidna, spiny anteater, anteater",
+    "103": "platypus, duckbill, duckbilled platypus, duck-billed platypus, Ornithorhynchus anatinus",
+    "104": "wallaby, brush kangaroo",
+    "105": "koala, koala bear, kangaroo bear, native bear, Phascolarctos cinereus",
+    "106": "wombat",
+    "107": "jellyfish",
+    "108": "sea anemone, anemone",
+    "109": "brain coral",
+    "110": "flatworm, platyhelminth",
+    "111": "nematode, nematode worm, roundworm",
+    "112": "conch",
+    "113": "snail",
+    "114": "slug",
+    "115": "sea slug, nudibranch",
+    "116": "chiton, coat-of-mail shell, sea cradle, polyplacophore",
+    "117": "chambered nautilus, pearly nautilus, nautilus",
+    "118": "Dungeness crab, Cancer magister",
+    "119": "rock crab, Cancer irroratus",
+    "120": "fiddler crab",
+    "121": "king crab, Alaska crab, Alaskan king crab, Alaska king crab, Paralithodes camtschatica",
+    "122": "American lobster, Northern lobster, Maine lobster, Homarus americanus",
+    "123": "spiny lobster, langouste, rock lobster, crawfish, crayfish, sea crawfish",
+    "124": "crayfish, crawfish, crawdad, crawdaddy",
+    "125": "hermit crab",
+    "126": "isopod",
+    "127": "white stork, Ciconia ciconia",
+    "128": "black stork, Ciconia nigra",
+    "129": "spoonbill",
+    "130": "flamingo",
+    "131": "little blue heron, Egretta caerulea",
+    "132": "American egret, great white heron, Egretta albus",
+    "133": "bittern",
+    "134": "crane",
+    "135": "limpkin, Aramus pictus",
+    "136": "European gallinule, Porphyrio porphyrio",
+    "137": "American coot, marsh hen, mud hen, water hen, Fulica americana",
+    "138": "bustard",
+    "139": "ruddy turnstone, Arenaria interpres",
+    "140": "red-backed sandpiper, dunlin, Erolia alpina",
+    "141": "redshank, Tringa totanus",
+    "142": "dowitcher",
+    "143": "oystercatcher, oyster catcher",
+    "144": "pelican",
+    "145": "king penguin, Aptenodytes patagonica",
+    "146": "albatross, mollymawk",
+    "147": "grey whale, gray whale, devilfish, Eschrichtius gibbosus, Eschrichtius robustus",
+    "148": "killer whale, killer, orca, grampus, sea wolf, Orcinus orca",
+    "149": "dugong, Dugong dugon",
+    "150": "sea lion",
+    "151": "Chihuahua",
+    "152": "Japanese spaniel",
+    "153": "Maltese dog, Maltese terrier, Maltese",
+    "154": "Pekinese, Pekingese, Peke",
+    "155": "Shih-Tzu",
+    "156": "Blenheim spaniel",
+    "157": "papillon",
+    "158": "toy terrier",
+    "159": "Rhodesian ridgeback",
+    "160": "Afghan hound, Afghan",
+    "161": "basset, basset hound",
+    "162": "beagle",
+    "163": "bloodhound, sleuthhound",
+    "164": "bluetick",
+    "165": "black-and-tan coonhound",
+    "166": "Walker hound, Walker foxhound",
+    "167": "English foxhound",
+    "168": "redbone",
+    "169": "borzoi, Russian wolfhound",
+    "170": "Irish wolfhound",
+    "171": "Italian greyhound",
+    "172": "whippet",
+    "173": "Ibizan hound, Ibizan Podenco",
+    "174": "Norwegian elkhound, elkhound",
+    "175": "otterhound, otter hound",
+    "176": "Saluki, gazelle hound",
+    "177": "Scottish deerhound, deerhound",
+    "178": "Weimaraner",
+    "179": "Staffordshire bullterrier, Staffordshire bull terrier",
+    "180": "American Staffordshire terrier, Staffordshire terrier, American pit bull terrier, pit bull terrier",
+    "181": "Bedlington terrier",
+    "182": "Border terrier",
+    "183": "Kerry blue terrier",
+    "184": "Irish terrier",
+    "185": "Norfolk terrier",
+    "186": "Norwich terrier",
+    "187": "Yorkshire terrier",
+    "188": "wire-haired fox terrier",
+    "189": "Lakeland terrier",
+    "190": "Sealyham terrier, Sealyham",
+    "191": "Airedale, Airedale terrier",
+    "192": "cairn, cairn terrier",
+    "193": "Australian terrier",
+    "194": "Dandie Dinmont, Dandie Dinmont terrier",
+    "195": "Boston bull, Boston terrier",
+    "196": "miniature schnauzer",
+    "197": "giant schnauzer",
+    "198": "standard schnauzer",
+    "199": "Scotch terrier, Scottish terrier, Scottie",
+    "200": "Tibetan terrier, chrysanthemum dog",
+    "201": "silky terrier, Sydney silky",
+    "202": "soft-coated wheaten terrier",
+    "203": "West Highland white terrier",
+    "204": "Lhasa, Lhasa apso",
+    "205": "flat-coated retriever",
+    "206": "curly-coated retriever",
+    "207": "golden retriever",
+    "208": "Labrador retriever",
+    "209": "Chesapeake Bay retriever",
+    "210": "German short-haired pointer",
+    "211": "vizsla, Hungarian pointer",
+    "212": "English setter",
+    "213": "Irish setter, red setter",
+    "214": "Gordon setter",
+    "215": "Brittany spaniel",
+    "216": "clumber, clumber spaniel",
+    "217": "English springer, English springer spaniel",
+    "218": "Welsh springer spaniel",
+    "219": "cocker spaniel, English cocker spaniel, cocker",
+    "220": "Sussex spaniel",
+    "221": "Irish water spaniel",
+    "222": "kuvasz",
+    "223": "schipperke",
+    "224": "groenendael",
+    "225": "malinois",
+    "226": "briard",
+    "227": "kelpie",
+    "228": "komondor",
+    "229": "Old English sheepdog, bobtail",
+    "230": "Shetland sheepdog, Shetland sheep dog, Shetland",
+    "231": "collie",
+    "232": "Border collie",
+    "233": "Bouvier des Flandres, Bouviers des Flandres",
+    "234": "Rottweiler",
+    "235": "German shepherd, German shepherd dog, German police dog, alsatian",
+    "236": "Doberman, Doberman pinscher",
+    "237": "miniature pinscher",
+    "238": "Greater Swiss Mountain dog",
+    "239": "Bernese mountain dog",
+    "240": "Appenzeller",
+    "241": "EntleBucher",
+    "242": "boxer",
+    "243": "bull mastiff",
+    "244": "Tibetan mastiff",
+    "245": "French bulldog",
+    "246": "Great Dane",
+    "247": "Saint Bernard, St Bernard",
+    "248": "Eskimo dog, husky",
+    "249": "malamute, malemute, Alaskan malamute",
+    "250": "Siberian husky",
+    "251": "dalmatian, coach dog, carriage dog",
+    "252": "affenpinscher, monkey pinscher, monkey dog",
+    "253": "basenji",
+    "254": "pug, pug-dog",
+    "255": "Leonberg",
+    "256": "Newfoundland, Newfoundland dog",
+    "257": "Great Pyrenees",
+    "258": "Samoyed, Samoyede",
+    "259": "Pomeranian",
+    "260": "chow, chow chow",
+    "261": "keeshond",
+    "262": "Brabancon griffon",
+    "263": "Pembroke, Pembroke Welsh corgi",
+    "264": "Cardigan, Cardigan Welsh corgi",
+    "265": "toy poodle",
+    "266": "miniature poodle",
+    "267": "standard poodle",
+    "268": "Mexican hairless",
+    "269": "timber wolf, grey wolf, gray wolf, Canis lupus",
+    "270": "white wolf, Arctic wolf, Canis lupus tundrarum",
+    "271": "red wolf, maned wolf, Canis rufus, Canis niger",
+    "272": "coyote, prairie wolf, brush wolf, Canis latrans",
+    "273": "dingo, warrigal, warragal, Canis dingo",
+    "274": "dhole, Cuon alpinus",
+    "275": "African hunting dog, hyena dog, Cape hunting dog, Lycaon pictus",
+    "276": "hyena, hyaena",
+    "277": "red fox, Vulpes vulpes",
+    "278": "kit fox, Vulpes macrotis",
+    "279": "Arctic fox, white fox, Alopex lagopus",
+    "280": "grey fox, gray fox, Urocyon cinereoargenteus",
+    "281": "tabby, tabby cat",
+    "282": "tiger cat",
+    "283": "Persian cat",
+    "284": "Siamese cat, Siamese",
+    "285": "Egyptian cat",
+    "286": "cougar, puma, catamount, mountain lion, painter, panther, Felis concolor",
+    "287": "lynx, catamount",
+    "288": "leopard, Panthera pardus",
+    "289": "snow leopard, ounce, Panthera uncia",
+    "290": "jaguar, panther, Panthera onca, Felis onca",
+    "291": "lion, king of beasts, Panthera leo",
+    "292": "tiger, Panthera tigris",
+    "293": "cheetah, chetah, Acinonyx jubatus",
+    "294": "brown bear, bruin, Ursus arctos",
+    "295": "American black bear, black bear, Ursus americanus, Euarctos americanus",
+    "296": "ice bear, polar bear, Ursus Maritimus, Thalarctos maritimus",
+    "297": "sloth bear, Melursus ursinus, Ursus ursinus",
+    "298": "mongoose",
+    "299": "meerkat, mierkat",
+    "300": "tiger beetle",
+    "301": "ladybug, ladybeetle, lady beetle, ladybird, ladybird beetle",
+    "302": "ground beetle, carabid beetle",
+    "303": "long-horned beetle, longicorn, longicorn beetle",
+    "304": "leaf beetle, chrysomelid",
+    "305": "dung beetle",
+    "306": "rhinoceros beetle",
+    "307": "weevil",
+    "308": "fly",
+    "309": "bee",
+    "310": "ant, emmet, pismire",
+    "311": "grasshopper, hopper",
+    "312": "cricket",
+    "313": "walking stick, walkingstick, stick insect",
+    "314": "cockroach, roach",
+    "315": "mantis, mantid",
+    "316": "cicada, cicala",
+    "317": "leafhopper",
+    "318": "lacewing, lacewing fly",
+    "319": "dragonfly, darning needle, devils darning needle, sewing needle, snake feeder, snake doctor, mosquito hawk, skeeter hawk",
+    "320": "damselfly",
+    "321": "admiral",
+    "322": "ringlet, ringlet butterfly",
+    "323": "monarch, monarch butterfly, milkweed butterfly, Danaus plexippus",
+    "324": "cabbage butterfly",
+    "325": "sulphur butterfly, sulfur butterfly",
+    "326": "lycaenid, lycaenid butterfly",
+    "327": "starfish, sea star",
+    "328": "sea urchin",
+    "329": "sea cucumber, holothurian",
+    "330": "wood rabbit, cottontail, cottontail rabbit",
+    "331": "hare",
+    "332": "Angora, Angora rabbit",
+    "333": "hamster",
+    "334": "porcupine, hedgehog",
+    "335": "fox squirrel, eastern fox squirrel, Sciurus niger",
+    "336": "marmot",
+    "337": "beaver",
+    "338": "guinea pig, Cavia cobaya",
+    "339": "sorrel",
+    "340": "zebra",
+    "341": "hog, pig, grunter, squealer, Sus scrofa",
+    "342": "wild boar, boar, Sus scrofa",
+    "343": "warthog",
+    "344": "hippopotamus, hippo, river horse, Hippopotamus amphibius",
+    "345": "ox",
+    "346": "water buffalo, water ox, Asiatic buffalo, Bubalus bubalis",
+    "347": "bison",
+    "348": "ram, tup",
+    "349": "bighorn, bighorn sheep, cimarron, Rocky Mountain bighorn, Rocky Mountain sheep, Ovis canadensis",
+    "350": "ibex, Capra ibex",
+    "351": "hartebeest",
+    "352": "impala, Aepyceros melampus",
+    "353": "gazelle",
+    "354": "Arabian camel, dromedary, Camelus dromedarius",
+    "355": "llama",
+    "356": "weasel",
+    "357": "mink",
+    "358": "polecat, fitch, foulmart, foumart, Mustela putorius",
+    "359": "black-footed ferret, ferret, Mustela nigripes",
+    "360": "otter",
+    "361": "skunk, polecat, wood pussy",
+    "362": "badger",
+    "363": "armadillo",
+    "364": "three-toed sloth, ai, Bradypus tridactylus",
+    "365": "orangutan, orang, orangutang, Pongo pygmaeus",
+    "366": "gorilla, Gorilla gorilla",
+    "367": "chimpanzee, chimp, Pan troglodytes",
+    "368": "gibbon, Hylobates lar",
+    "369": "siamang, Hylobates syndactylus, Symphalangus syndactylus",
+    "370": "guenon, guenon monkey",
+    "371": "patas, hussar monkey, Erythrocebus patas",
+    "372": "baboon",
+    "373": "macaque",
+    "374": "langur",
+    "375": "colobus, colobus monkey",
+    "376": "proboscis monkey, Nasalis larvatus",
+    "377": "marmoset",
+    "378": "capuchin, ringtail, Cebus capucinus",
+    "379": "howler monkey, howler",
+    "380": "titi, titi monkey",
+    "381": "spider monkey, Ateles geoffroyi",
+    "382": "squirrel monkey, Saimiri sciureus",
+    "383": "Madagascar cat, ring-tailed lemur, Lemur catta",
+    "384": "indri, indris, Indri indri, Indri brevicaudatus",
+    "385": "Indian elephant, Elephas maximus",
+    "386": "African elephant, Loxodonta africana",
+    "387": "lesser panda, red panda, panda, bear cat, cat bear, Ailurus fulgens",
+    "388": "giant panda, panda, panda bear, coon bear, Ailuropoda melanoleuca",
+    "389": "barracouta, snoek",
+    "390": "eel",
+    "391": "coho, cohoe, coho salmon, blue jack, silver salmon, Oncorhynchus kisutch",
+    "392": "rock beauty, Holocanthus tricolor",
+    "393": "anemone fish",
+    "394": "sturgeon",
+    "395": "gar, garfish, garpike, billfish, Lepisosteus osseus",
+    "396": "lionfish",
+    "397": "puffer, pufferfish, blowfish, globefish",
+    "398": "abacus",
+    "399": "abaya",
+    "400": "academic gown, academic robe, judge robe",
+    "401": "accordion, piano accordion, squeeze box",
+    "402": "acoustic guitar",
+    "403": "aircraft carrier, carrier, flattop, attack aircraft carrier",
+    "404": "airliner",
+    "405": "airship, dirigible",
+    "406": "altar",
+    "407": "ambulance",
+    "408": "amphibian, amphibious vehicle",
+    "409": "analog clock",
+    "410": "apiary, bee house",
+    "411": "apron",
+    "412": "ashcan, trash can, garbage can, wastebin, ash bin, ash-bin, ashbin, dustbin, trash barrel, trash bin",
+    "413": "assault rifle, assault gun",
+    "414": "backpack, back pack, knapsack, packsack, rucksack, haversack",
+    "415": "bakery, bakeshop, bakehouse",
+    "416": "balance beam, beam",
+    "417": "balloon",
+    "418": "ballpoint, ballpoint pen, ballpen, Biro",
+    "419": "Band Aid",
+    "420": "banjo",
+    "421": "bannister, banister, balustrade, balusters, handrail",
+    "422": "barbell",
+    "423": "barber chair",
+    "424": "barbershop",
+    "425": "barn",
+    "426": "barometer",
+    "427": "barrel, cask",
+    "428": "barrow, garden cart, lawn cart, wheelbarrow",
+    "429": "baseball",
+    "430": "basketball",
+    "431": "bassinet",
+    "432": "bassoon",
+    "433": "bathing cap, swimming cap",
+    "434": "bath towel",
+    "435": "bathtub, bathing tub, bath, tub",
+    "436": "beach wagon, station wagon, wagon, estate car, beach waggon, station waggon, waggon",
+    "437": "beacon, lighthouse, beacon light, pharos",
+    "438": "beaker",
+    "439": "bearskin, busby, shako",
+    "440": "beer bottle",
+    "441": "beer glass",
+    "442": "bell cote, bell cot",
+    "443": "bib",
+    "444": "bicycle-built-for-two, tandem bicycle, tandem",
+    "445": "bikini, two-piece",
+    "446": "binder, ring-binder",
+    "447": "binoculars, field glasses, opera glasses",
+    "448": "birdhouse",
+    "449": "boathouse",
+    "450": "bobsled, bobsleigh, bob",
+    "451": "bolo tie, bolo, bola tie, bola",
+    "452": "bonnet, poke bonnet",
+    "453": "bookcase",
+    "454": "bookshop, bookstore, bookstall",
+    "455": "bottlecap",
+    "456": "bow",
+    "457": "bow tie, bow-tie, bowtie",
+    "458": "brass, memorial tablet, plaque",
+    "459": "brassiere, bra, bandeau",
+    "460": "breakwater, groin, groyne, mole, bulwark, seawall, jetty",
+    "461": "breastplate, aegis, egis",
+    "462": "broom",
+    "463": "bucket, pail",
+    "464": "buckle",
+    "465": "bulletproof vest",
+    "466": "bullet train, bullet",
+    "467": "butcher shop, meat market",
+    "468": "cab, hack, taxi, taxicab",
+    "469": "caldron, cauldron",
+    "470": "candle, taper, wax light",
+    "471": "cannon",
+    "472": "canoe",
+    "473": "can opener, tin opener",
+    "474": "cardigan",
+    "475": "car mirror",
+    "476": "carousel, carrousel, merry-go-round, roundabout, whirligig",
+    "477": "carpenters kit, tool kit",
+    "478": "carton",
+    "479": "car wheel",
+    "480": "cash machine, cash dispenser, automated teller machine, automatic teller machine, automated teller, automatic teller, ATM",
+    "481": "cassette",
+    "482": "cassette player",
+    "483": "castle",
+    "484": "catamaran",
+    "485": "CD player",
+    "486": "cello, violoncello",
+    "487": "cellular telephone, cellular phone, cellphone, cell, mobile phone",
+    "488": "chain",
+    "489": "chainlink fence",
+    "490": "chain mail, ring mail, mail, chain armor, chain armour, ring armor, ring armour",
+    "491": "chain saw, chainsaw",
+    "492": "chest",
+    "493": "chiffonier, commode",
+    "494": "chime, bell, gong",
+    "495": "china cabinet, china closet",
+    "496": "Christmas stocking",
+    "497": "church, church building",
+    "498": "cinema, movie theater, movie theatre, movie house, picture palace",
+    "499": "cleaver, meat cleaver, chopper",
+    "500": "cliff dwelling",
+    "501": "cloak",
+    "502": "clog, geta, patten, sabot",
+    "503": "cocktail shaker",
+    "504": "coffee mug",
+    "505": "coffeepot",
+    "506": "coil, spiral, volute, whorl, helix",
+    "507": "combination lock",
+    "508": "computer keyboard, keypad",
+    "509": "confectionery, confectionary, candy store",
+    "510": "container ship, containership, container vessel",
+    "511": "convertible",
+    "512": "corkscrew, bottle screw",
+    "513": "cornet, horn, trumpet, trump",
+    "514": "cowboy boot",
+    "515": "cowboy hat, ten-gallon hat",
+    "516": "cradle",
+    "517": "crane",
+    "518": "crash helmet",
+    "519": "crate",
+    "520": "crib, cot",
+    "521": "Crock Pot",
+    "522": "croquet ball",
+    "523": "crutch",
+    "524": "cuirass",
+    "525": "dam, dike, dyke",
+    "526": "desk",
+    "527": "desktop computer",
+    "528": "dial telephone, dial phone",
+    "529": "diaper, nappy, napkin",
+    "530": "digital clock",
+    "531": "digital watch",
+    "532": "dining table, board",
+    "533": "dishrag, dishcloth",
+    "534": "dishwasher, dish washer, dishwashing machine",
+    "535": "disk brake, disc brake",
+    "536": "dock, dockage, docking facility",
+    "537": "dogsled, dog sled, dog sleigh",
+    "538": "dome",
+    "539": "doormat, welcome mat",
+    "540": "drilling platform, offshore rig",
+    "541": "drum, membranophone, tympan",
+    "542": "drumstick",
+    "543": "dumbbell",
+    "544": "Dutch oven",
+    "545": "electric fan, blower",
+    "546": "electric guitar",
+    "547": "electric locomotive",
+    "548": "entertainment center",
+    "549": "envelope",
+    "550": "espresso maker",
+    "551": "face powder",
+    "552": "feather boa, boa",
+    "553": "file, file cabinet, filing cabinet",
+    "554": "fireboat",
+    "555": "fire engine, fire truck",
+    "556": "fire screen, fireguard",
+    "557": "flagpole, flagstaff",
+    "558": "flute, transverse flute",
+    "559": "folding chair",
+    "560": "football helmet",
+    "561": "forklift",
+    "562": "fountain",
+    "563": "fountain pen",
+    "564": "four-poster",
+    "565": "freight car",
+    "566": "French horn, horn",
+    "567": "frying pan, frypan, skillet",
+    "568": "fur coat",
+    "569": "garbage truck, dustcart",
+    "570": "gasmask, respirator, gas helmet",
+    "571": "gas pump, gasoline pump, petrol pump, island dispenser",
+    "572": "goblet",
+    "573": "go-kart",
+    "574": "golf ball",
+    "575": "golfcart, golf cart",
+    "576": "gondola",
+    "577": "gong, tam-tam",
+    "578": "gown",
+    "579": "grand piano, grand",
+    "580": "greenhouse, nursery, glasshouse",
+    "581": "grille, radiator grille",
+    "582": "grocery store, grocery, food market, market",
+    "583": "guillotine",
+    "584": "hair slide",
+    "585": "hair spray",
+    "586": "half track",
+    "587": "hammer",
+    "588": "hamper",
+    "589": "hand blower, blow dryer, blow drier, hair dryer, hair drier",
+    "590": "hand-held computer, hand-held microcomputer",
+    "591": "handkerchief, hankie, hanky, hankey",
+    "592": "hard disc, hard disk, fixed disk",
+    "593": "harmonica, mouth organ, harp, mouth harp",
+    "594": "harp",
+    "595": "harvester, reaper",
+    "596": "hatchet",
+    "597": "holster",
+    "598": "home theater, home theatre",
+    "599": "honeycomb",
+    "600": "hook, claw",
+    "601": "hoopskirt, crinoline",
+    "602": "horizontal bar, high bar",
+    "603": "horse cart, horse-cart",
+    "604": "hourglass",
+    "605": "iPod",
+    "606": "iron, smoothing iron",
+    "607": "jack-o-lantern",
+    "608": "jean, blue jean, denim",
+    "609": "jeep, landrover",
+    "610": "jersey, T-shirt, tee shirt",
+    "611": "jigsaw puzzle",
+    "612": "jinrikisha, ricksha, rickshaw",
+    "613": "joystick",
+    "614": "kimono",
+    "615": "knee pad",
+    "616": "knot",
+    "617": "lab coat, laboratory coat",
+    "618": "ladle",
+    "619": "lampshade, lamp shade",
+    "620": "laptop, laptop computer",
+    "621": "lawn mower, mower",
+    "622": "lens cap, lens cover",
+    "623": "letter opener, paper knife, paperknife",
+    "624": "library",
+    "625": "lifeboat",
+    "626": "lighter, light, igniter, ignitor",
+    "627": "limousine, limo",
+    "628": "liner, ocean liner",
+    "629": "lipstick, lip rouge",
+    "630": "Loafer",
+    "631": "lotion",
+    "632": "loudspeaker, speaker, speaker unit, loudspeaker system, speaker system",
+    "633": "loupe, jewelers loupe",
+    "634": "lumbermill, sawmill",
+    "635": "magnetic compass",
+    "636": "mailbag, postbag",
+    "637": "mailbox, letter box",
+    "638": "maillot",
+    "639": "maillot, tank suit",
+    "640": "manhole cover",
+    "641": "maraca",
+    "642": "marimba, xylophone",
+    "643": "mask",
+    "644": "matchstick",
+    "645": "maypole",
+    "646": "maze, labyrinth",
+    "647": "measuring cup",
+    "648": "medicine chest, medicine cabinet",
+    "649": "megalith, megalithic structure",
+    "650": "microphone, mike",
+    "651": "microwave, microwave oven",
+    "652": "military uniform",
+    "653": "milk can",
+    "654": "minibus",
+    "655": "miniskirt, mini",
+    "656": "minivan",
+    "657": "missile",
+    "658": "mitten",
+    "659": "mixing bowl",
+    "660": "mobile home, manufactured home",
+    "661": "Model T",
+    "662": "modem",
+    "663": "monastery",
+    "664": "monitor",
+    "665": "moped",
+    "666": "mortar",
+    "667": "mortarboard",
+    "668": "mosque",
+    "669": "mosquito net",
+    "670": "motor scooter, scooter",
+    "671": "mountain bike, all-terrain bike, off-roader",
+    "672": "mountain tent",
+    "673": "mouse, computer mouse",
+    "674": "mousetrap",
+    "675": "moving van",
+    "676": "muzzle",
+    "677": "nail",
+    "678": "neck brace",
+    "679": "necklace",
+    "680": "nipple",
+    "681": "notebook, notebook computer",
+    "682": "obelisk",
+    "683": "oboe, hautboy, hautbois",
+    "684": "ocarina, sweet potato",
+    "685": "odometer, hodometer, mileometer, milometer",
+    "686": "oil filter",
+    "687": "organ, pipe organ",
+    "688": "oscilloscope, scope, cathode-ray oscilloscope, CRO",
+    "689": "overskirt",
+    "690": "oxcart",
+    "691": "oxygen mask",
+    "692": "packet",
+    "693": "paddle, boat paddle",
+    "694": "paddlewheel, paddle wheel",
+    "695": "padlock",
+    "696": "paintbrush",
+    "697": "pajama, pyjama, pjs, jammies",
+    "698": "palace",
+    "699": "panpipe, pandean pipe, syrinx",
+    "700": "paper towel",
+    "701": "parachute, chute",
+    "702": "parallel bars, bars",
+    "703": "park bench",
+    "704": "parking meter",
+    "705": "passenger car, coach, carriage",
+    "706": "patio, terrace",
+    "707": "pay-phone, pay-station",
+    "708": "pedestal, plinth, footstall",
+    "709": "pencil box, pencil case",
+    "710": "pencil sharpener",
+    "711": "perfume, essence",
+    "712": "Petri dish",
+    "713": "photocopier",
+    "714": "pick, plectrum, plectron",
+    "715": "pickelhaube",
+    "716": "picket fence, paling",
+    "717": "pickup, pickup truck",
+    "718": "pier",
+    "719": "piggy bank, penny bank",
+    "720": "pill bottle",
+    "721": "pillow",
+    "722": "ping-pong ball",
+    "723": "pinwheel",
+    "724": "pirate, pirate ship",
+    "725": "pitcher, ewer",
+    "726": "plane, carpenters plane, woodworking plane",
+    "727": "planetarium",
+    "728": "plastic bag",
+    "729": "plate rack",
+    "730": "plow, plough",
+    "731": "plunger, plumbers helper",
+    "732": "Polaroid camera, Polaroid Land camera",
+    "733": "pole",
+    "734": "police van, police wagon, paddy wagon, patrol wagon, wagon, black Maria",
+    "735": "poncho",
+    "736": "pool table, billiard table, snooker table",
+    "737": "pop bottle, soda bottle",
+    "738": "pot, flowerpot",
+    "739": "potters wheel",
+    "740": "power drill",
+    "741": "prayer rug, prayer mat",
+    "742": "printer",
+    "743": "prison, prison house",
+    "744": "projectile, missile",
+    "745": "projector",
+    "746": "puck, hockey puck",
+    "747": "punching bag, punch bag, punching ball, punchball",
+    "748": "purse",
+    "749": "quill, quill pen",
+    "750": "quilt, comforter, comfort, puff",
+    "751": "racer, race car, racing car",
+    "752": "racket, racquet",
+    "753": "radiator",
+    "754": "radio, wireless",
+    "755": "radio telescope, radio reflector",
+    "756": "rain barrel",
+    "757": "recreational vehicle, RV, R.V.",
+    "758": "reel",
+    "759": "reflex camera",
+    "760": "refrigerator, icebox",
+    "761": "remote control, remote",
+    "762": "restaurant, eating house, eating place, eatery",
+    "763": "revolver, six-gun, six-shooter",
+    "764": "rifle",
+    "765": "rocking chair, rocker",
+    "766": "rotisserie",
+    "767": "rubber eraser, rubber, pencil eraser",
+    "768": "rugby ball",
+    "769": "rule, ruler",
+    "770": "running shoe",
+    "771": "safe",
+    "772": "safety pin",
+    "773": "saltshaker, salt shaker",
+    "774": "sandal",
+    "775": "sarong",
+    "776": "sax, saxophone",
+    "777": "scabbard",
+    "778": "scale, weighing machine",
+    "779": "school bus",
+    "780": "schooner",
+    "781": "scoreboard",
+    "782": "screen, CRT screen",
+    "783": "screw",
+    "784": "screwdriver",
+    "785": "seat belt, seatbelt",
+    "786": "sewing machine",
+    "787": "shield, buckler",
+    "788": "shoe shop, shoe-shop, shoe store",
+    "789": "shoji",
+    "790": "shopping basket",
+    "791": "shopping cart",
+    "792": "shovel",
+    "793": "shower cap",
+    "794": "shower curtain",
+    "795": "ski",
+    "796": "ski mask",
+    "797": "sleeping bag",
+    "798": "slide rule, slipstick",
+    "799": "sliding door",
+    "800": "slot, one-armed bandit",
+    "801": "snorkel",
+    "802": "snowmobile",
+    "803": "snowplow, snowplough",
+    "804": "soap dispenser",
+    "805": "soccer ball",
+    "806": "sock",
+    "807": "solar dish, solar collector, solar furnace",
+    "808": "sombrero",
+    "809": "soup bowl",
+    "810": "space bar",
+    "811": "space heater",
+    "812": "space shuttle",
+    "813": "spatula",
+    "814": "speedboat",
+    "815": "spider web, spiders web",
+    "816": "spindle",
+    "817": "sports car, sport car",
+    "818": "spotlight, spot",
+    "819": "stage",
+    "820": "steam locomotive",
+    "821": "steel arch bridge",
+    "822": "steel drum",
+    "823": "stethoscope",
+    "824": "stole",
+    "825": "stone wall",
+    "826": "stopwatch, stop watch",
+    "827": "stove",
+    "828": "strainer",
+    "829": "streetcar, tram, tramcar, trolley, trolley car",
+    "830": "stretcher",
+    "831": "studio couch, day bed",
+    "832": "stupa, tope",
+    "833": "submarine, pigboat, sub, U-boat",
+    "834": "suit, suit of clothes",
+    "835": "sundial",
+    "836": "sunglass",
+    "837": "sunglasses, dark glasses, shades",
+    "838": "sunscreen, sunblock, sun blocker",
+    "839": "suspension bridge",
+    "840": "swab, swob, mop",
+    "841": "sweatshirt",
+    "842": "swimming trunks, bathing trunks",
+    "843": "swing",
+    "844": "switch, electric switch, electrical switch",
+    "845": "syringe",
+    "846": "table lamp",
+    "847": "tank, army tank, armored combat vehicle, armoured combat vehicle",
+    "848": "tape player",
+    "849": "teapot",
+    "850": "teddy, teddy bear",
+    "851": "television, television system",
+    "852": "tennis ball",
+    "853": "thatch, thatched roof",
+    "854": "theater curtain, theatre curtain",
+    "855": "thimble",
+    "856": "thresher, thrasher, threshing machine",
+    "857": "throne",
+    "858": "tile roof",
+    "859": "toaster",
+    "860": "tobacco shop, tobacconist shop, tobacconist",
+    "861": "toilet seat",
+    "862": "torch",
+    "863": "totem pole",
+    "864": "tow truck, tow car, wrecker",
+    "865": "toyshop",
+    "866": "tractor",
+    "867": "trailer truck, tractor trailer, trucking rig, rig, articulated lorry, semi",
+    "868": "tray",
+    "869": "trench coat",
+    "870": "tricycle, trike, velocipede",
+    "871": "trimaran",
+    "872": "tripod",
+    "873": "triumphal arch",
+    "874": "trolleybus, trolley coach, trackless trolley",
+    "875": "trombone",
+    "876": "tub, vat",
+    "877": "turnstile",
+    "878": "typewriter keyboard",
+    "879": "umbrella",
+    "880": "unicycle, monocycle",
+    "881": "upright, upright piano",
+    "882": "vacuum, vacuum cleaner",
+    "883": "vase",
+    "884": "vault",
+    "885": "velvet",
+    "886": "vending machine",
+    "887": "vestment",
+    "888": "viaduct",
+    "889": "violin, fiddle",
+    "890": "volleyball",
+    "891": "waffle iron",
+    "892": "wall clock",
+    "893": "wallet, billfold, notecase, pocketbook",
+    "894": "wardrobe, closet, press",
+    "895": "warplane, military plane",
+    "896": "washbasin, handbasin, washbowl, lavabo, wash-hand basin",
+    "897": "washer, automatic washer, washing machine",
+    "898": "water bottle",
+    "899": "water jug",
+    "900": "water tower",
+    "901": "whiskey jug",
+    "902": "whistle",
+    "903": "wig",
+    "904": "window screen",
+    "905": "window shade",
+    "906": "Windsor tie",
+    "907": "wine bottle",
+    "908": "wing",
+    "909": "wok",
+    "910": "wooden spoon",
+    "911": "wool, woolen, woollen",
+    "912": "worm fence, snake fence, snake-rail fence, Virginia fence",
+    "913": "wreck",
+    "914": "yawl",
+    "915": "yurt",
+    "916": "web site, website, internet site, site",
+    "917": "comic book",
+    "918": "crossword puzzle, crossword",
+    "919": "street sign",
+    "920": "traffic light, traffic signal, stoplight",
+    "921": "book jacket, dust cover, dust jacket, dust wrapper",
+    "922": "menu",
+    "923": "plate",
+    "924": "guacamole",
+    "925": "consomme",
+    "926": "hot pot, hotpot",
+    "927": "trifle",
+    "928": "ice cream, icecream",
+    "929": "ice lolly, lolly, lollipop, popsicle",
+    "930": "French loaf",
+    "931": "bagel, beigel",
+    "932": "pretzel",
+    "933": "cheeseburger",
+    "934": "hotdog, hot dog, red hot",
+    "935": "mashed potato",
+    "936": "head cabbage",
+    "937": "broccoli",
+    "938": "cauliflower",
+    "939": "zucchini, courgette",
+    "940": "spaghetti squash",
+    "941": "acorn squash",
+    "942": "butternut squash",
+    "943": "cucumber, cuke",
+    "944": "artichoke, globe artichoke",
+    "945": "bell pepper",
+    "946": "cardoon",
+    "947": "mushroom",
+    "948": "Granny Smith",
+    "949": "strawberry",
+    "950": "orange",
+    "951": "lemon",
+    "952": "fig",
+    "953": "pineapple, ananas",
+    "954": "banana",
+    "955": "jackfruit, jak, jack",
+    "956": "custard apple",
+    "957": "pomegranate",
+    "958": "hay",
+    "959": "carbonara",
+    "960": "chocolate sauce, chocolate syrup",
+    "961": "dough",
+    "962": "meat loaf, meatloaf",
+    "963": "pizza, pizza pie",
+    "964": "potpie",
+    "965": "burrito",
+    "966": "red wine",
+    "967": "espresso",
+    "968": "cup",
+    "969": "eggnog",
+    "970": "alp",
+    "971": "bubble",
+    "972": "cliff, drop, drop-off",
+    "973": "coral reef",
+    "974": "geyser",
+    "975": "lakeside, lakeshore",
+    "976": "promontory, headland, head, foreland",
+    "977": "sandbar, sand bar",
+    "978": "seashore, coast, seacoast, sea-coast",
+    "979": "valley, vale",
+    "980": "volcano",
+    "981": "ballplayer, baseball player",
+    "982": "groom, bridegroom",
+    "983": "scuba diver",
+    "984": "rapeseed",
+    "985": "daisy",
+    "986": "yellow ladys slipper, yellow lady-slipper, Cypripedium calceolus, Cypripedium parviflorum",
+    "987": "corn",
+    "988": "acorn",
+    "989": "hip, rose hip, rosehip",
+    "990": "buckeye, horse chestnut, conker",
+    "991": "coral fungus",
+    "992": "agaric",
+    "993": "gyromitra",
+    "994": "stinkhorn, carrion fungus",
+    "995": "earthstar",
+    "996": "hen-of-the-woods, hen of the woods, Polyporus frondosus, Grifola frondosa",
+    "997": "bolete",
+    "998": "ear, spike, capitulum",
+    "999": "toilet tissue, toilet paper, bathroom tissue"
+  }
+}

iMF-L-2/pipeline.py

@@ -0,0 +1,275 @@
+# Copyright 2026 The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+"""Hub custom pipeline: IMFPipeline.
+
+Load with native Hugging Face diffusers and trust_remote_code=True.
+"""
+
+from __future__ import annotations
+
+import json
+from pathlib import Path
+from typing import Dict, List, Optional, Tuple, Union
+
+import torch
+from diffusers.image_processor import VaeImageProcessor
+from diffusers.pipelines.pipeline_utils import DiffusionPipeline, ImagePipelineOutput
+from diffusers.schedulers import FlowMatchEulerDiscreteScheduler
+from diffusers.utils.torch_utils import randn_tensor
+
+
+def _set_imf_timesteps(
+    scheduler: FlowMatchEulerDiscreteScheduler,
+    num_inference_steps: int,
+    device: torch.device,
+) -> torch.Tensor:
+    flow_sigmas = torch.linspace(1.0, 0.0, num_inference_steps + 1, device=device, dtype=torch.float32)
+    scheduler.set_timesteps(sigmas=flow_sigmas.tolist(), device=device)
+    return flow_sigmas
+
+
+class IMFPipeline(DiffusionPipeline):
+    r"""
+    Pipeline for ImageNet class-conditional generation with Improved Mean Flows (iMF).
+
+    Parameters:
+        transformer ([`IMFTransformer2DModel`]):
+            Class-conditioned iMF transformer that predicts mean-flow velocity.
+        scheduler ([`FlowMatchEulerDiscreteScheduler`]):
+            Flow-matching Euler scheduler.
+        vae ([`AutoencoderKL`]):
+            Variational autoencoder used to decode transformer latents to pixels.
+        id2label (`dict[int, str]`, *optional*):
+            ImageNet class id to English label mapping.
+    """
+
+    model_cpu_offload_seq = "transformer->vae"
+
+    def __init__(
+        self,
+        transformer,
+        scheduler,
+        vae,
+        id2label: Optional[Dict[Union[int, str], str]] = None,
+    ):
+        super().__init__()
+        if scheduler is None:
+            scheduler = FlowMatchEulerDiscreteScheduler(
+                num_train_timesteps=1000,
+                shift=1.0,
+                stochastic_sampling=False,
+            )
+        self.register_modules(transformer=transformer, scheduler=scheduler, vae=vae)
+        self.vae_scale_factor = 2 ** (len(self.vae.config.block_out_channels) - 1)
+        self.image_processor = VaeImageProcessor(vae_scale_factor=self.vae_scale_factor)
+        self._id2label = self._normalize_id2label(id2label)
+        self.labels = self._build_label2id(self._id2label)
+        self._labels_loaded_from_model_index = bool(self._id2label)
+
+    def _ensure_labels_loaded(self) -> None:
+        if self._labels_loaded_from_model_index:
+            return
+        loaded = self._read_id2label_from_model_index(getattr(self.config, "_name_or_path", None))
+        if loaded:
+            self._id2label = loaded
+            self.labels = self._build_label2id(self._id2label)
+        self._labels_loaded_from_model_index = True
+
+    @staticmethod
+    def _normalize_id2label(id2label: Optional[Dict[Union[int, str], str]]) -> Dict[int, str]:
+        if not id2label:
+            return {}
+        return {int(key): value for key, value in id2label.items()}
+
+    @staticmethod
+    def _read_id2label_from_model_index(variant_path: Optional[str]) -> Dict[int, str]:
+        if not variant_path:
+            return {}
+        model_index_path = Path(variant_path).resolve() / "model_index.json"
+        if not model_index_path.exists():
+            return {}
+        raw = json.loads(model_index_path.read_text(encoding="utf-8"))
+        id2label = raw.get("id2label")
+        if not isinstance(id2label, dict):
+            return {}
+        return {int(key): value for key, value in id2label.items()}
+
+    @staticmethod
+    def _build_label2id(id2label: Dict[int, str]) -> Dict[str, int]:
+        label2id: Dict[str, int] = {}
+        for class_id, value in id2label.items():
+            for synonym in value.split(","):
+                synonym = synonym.strip()
+                if synonym:
+                    label2id[synonym] = int(class_id)
+        return dict(sorted(label2id.items()))
+
+    @property
+    def id2label(self) -> Dict[int, str]:
+        self._ensure_labels_loaded()
+        return self._id2label
+
+    def get_label_ids(self, label: Union[str, List[str]]) -> List[int]:
+        self._ensure_labels_loaded()
+        if not self.labels:
+            raise ValueError("No labels loaded. Ensure `id2label` exists in model_index.json.")
+        labels = [label] if isinstance(label, str) else label
+        missing = [item for item in labels if item not in self.labels]
+        if missing:
+            preview = ", ".join(list(self.labels.keys())[:8])
+            raise ValueError(f"Unknown label(s): {missing}. Example valid labels: {preview}, ...")
+        return [self.labels[item] for item in labels]
+
+    def _normalize_class_labels(self, class_labels: Union[int, str, List[Union[int, str]]]) -> List[int]:
+        if isinstance(class_labels, int):
+            return [class_labels]
+        if isinstance(class_labels, str):
+            return self.get_label_ids(class_labels)
+        if class_labels and isinstance(class_labels[0], str):
+            return self.get_label_ids(class_labels)
+        return list(class_labels)
+
+    def _predict_velocity_u(
+        self,
+        latents: torch.Tensor,
+        timestep: torch.Tensor,
+        time_gap: torch.Tensor,
+        class_labels: torch.Tensor,
+        class_null: torch.Tensor,
+        guidance_scale: float,
+        guidance_interval_start: float,
+        guidance_interval_end: float,
+        do_classifier_free_guidance: bool,
+    ) -> torch.Tensor:
+        dtype = latents.dtype
+        timestep = timestep.to(device=latents.device, dtype=dtype)
+        time_gap = time_gap.to(device=latents.device, dtype=dtype)
+
+        if do_classifier_free_guidance:
+            latents_in = torch.cat([latents, latents], dim=0)
+            labels = torch.cat([class_labels, class_null], dim=0)
+            omega = torch.tensor([guidance_scale, 1.0], device=latents.device, dtype=dtype)
+            t_min = torch.tensor([guidance_interval_start, 0.0], device=latents.device, dtype=dtype)
+            t_max = torch.tensor([guidance_interval_end, 1.0], device=latents.device, dtype=dtype)
+            batch = latents.shape[0]
+            timestep_in = timestep.reshape(1).repeat(2 * batch)
+            time_gap_in = time_gap.reshape(1).repeat(2 * batch)
+            omega = omega.repeat(batch)
+            t_min = t_min.repeat(batch)
+            t_max = t_max.repeat(batch)
+        else:
+            latents_in = latents
+            labels = class_labels
+            batch = latents.shape[0]
+            timestep_in = timestep.reshape(1).repeat(batch)
+            time_gap_in = time_gap.reshape(1).repeat(batch)
+            omega = torch.full((batch,), guidance_scale, device=latents.device, dtype=dtype)
+            t_min = torch.full((batch,), guidance_interval_start, device=latents.device, dtype=dtype)
+            t_max = torch.full((batch,), guidance_interval_end, device=latents.device, dtype=dtype)
+
+        outputs = self.transformer(
+            sample=latents_in,
+            timestep=timestep_in,
+            class_labels=labels,
+            time_gap=time_gap_in,
+            guidance_scale=omega,
+            guidance_interval_start=t_min,
+            guidance_interval_end=t_max,
+            return_dict=True,
+        )
+        velocity_u = outputs.velocity_u
+
+        if not do_classifier_free_guidance:
+            return velocity_u
+
+        u_cond, u_uncond = velocity_u.chunk(2, dim=0)
+        return u_uncond + guidance_scale * (u_cond - u_uncond)
+
+    def decode_latents(self, latents: torch.Tensor, output_type: str = "pil"):
+        if output_type == "latent":
+            return latents
+
+        scaling_factor = self.vae.config.scaling_factor
+        latents = latents.to(device=self.vae.device, dtype=self.vae.dtype)
+        image = self.vae.decode(latents / scaling_factor).sample
+        if output_type == "pt":
+            return image
+        return self.image_processor.postprocess(image, output_type=output_type)
+
+    @torch.inference_mode()
+    def __call__(
+        self,
+        class_labels: Union[int, str, List[Union[int, str]]],
+        num_inference_steps: int = 1,
+        guidance_scale: float = 2.7,
+        guidance_interval_start: float = 0.1,
+        guidance_interval_end: float = 0.9,
+        generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
+        latents: Optional[torch.Tensor] = None,
+        output_type: str = "pil",
+        return_dict: bool = True,
+    ) -> Union[ImagePipelineOutput, Tuple]:
+        if output_type not in {"pil", "np", "pt", "latent"}:
+            raise ValueError("output_type must be one of: 'pil', 'np', 'pt', 'latent'.")
+
+        class_label_ids = self._normalize_class_labels(class_labels)
+        do_classifier_free_guidance = guidance_scale > 1.0
+        batch_size = len(class_label_ids)
+
+        image_size = int(self.transformer.config.sample_size)
+        channels = int(self.transformer.config.in_channels)
+        null_class_val = int(self.transformer.config.num_classes)
+
+        if latents is None:
+            latents = randn_tensor(
+                shape=(batch_size, channels, image_size, image_size),
+                generator=generator,
+                device=self._execution_device,
+                dtype=self.transformer.dtype,
+            )
+
+        class_labels_t = torch.tensor(class_label_ids, device=latents.device, dtype=torch.long).reshape(-1)
+        class_labels_t = class_labels_t.clamp(0, null_class_val - 1)
+        class_null = torch.full_like(class_labels_t, null_class_val)
+
+        flow_sigmas = _set_imf_timesteps(self.scheduler, num_inference_steps, latents.device)
+
+        for i in self.progress_bar(range(num_inference_steps)):
+            t = flow_sigmas[i]
+            t_next = flow_sigmas[i + 1]
+            time_gap = t - t_next
+            velocity_u = self._predict_velocity_u(
+                latents,
+                t,
+                time_gap,
+                class_labels_t,
+                class_null,
+                guidance_scale,
+                guidance_interval_start,
+                guidance_interval_end,
+                do_classifier_free_guidance,
+            )
+            latents = self.scheduler.step(velocity_u, self.scheduler.timesteps[i], latents).prev_sample
+
+        images = self.decode_latents(latents, output_type=output_type)
+
+        self.maybe_free_model_hooks()
+
+        if not return_dict:
+            return (images,)
+        return ImagePipelineOutput(images=images)
+
+
+IMFPipelineOutput = ImagePipelineOutput

iMF-L-2/scheduler/scheduler_config.json

@@ -0,0 +1,7 @@
+{
+  "_class_name": "FlowMatchEulerDiscreteScheduler",
+  "_diffusers_version": "0.36.0",
+  "num_train_timesteps": 1000,
+  "shift": 1.0,
+  "stochastic_sampling": false
+}

iMF-L-2/transformer/config.json

@@ -0,0 +1,23 @@
+{
+  "_class_name": "IMFTransformer2DModel",
+  "_diffusers_version": "0.38.0",
+  "aux_head_depth": 8,
+  "depth": 32,
+  "embedding_init_constant": 1.0,
+  "eval_mode": true,
+  "hidden_size": 1024,
+  "in_channels": 4,
+  "mlp_ratio": 2.6666666666666665,
+  "model_type": "iMF-L/2",
+  "num_attention_heads": 16,
+  "num_cfg_tokens": 4,
+  "num_class_embeds": null,
+  "num_class_tokens": 8,
+  "num_classes": 1000,
+  "num_interval_tokens": 2,
+  "num_time_tokens": 4,
+  "patch_size": 2,
+  "sample_size": 32,
+  "token_init_constant": 1.0,
+  "weight_init_constant": 0.32
+}

iMF-L-2/transformer/diffusion_pytorch_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7157ae20a227965cd82e8e91d1359886d674bed63d03213fd4fe8e6a47a1c6fe
+size 1636334400

iMF-L-2/transformer/transformer_imf.py

@@ -0,0 +1,586 @@
+from __future__ import annotations
+
+import argparse
+import math
+from collections.abc import Mapping
+from dataclasses import dataclass
+from functools import partial
+from math import sqrt
+from typing import Dict, Literal, Optional, Tuple, Union
+
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+from diffusers.configuration_utils import ConfigMixin, register_to_config
+from diffusers.models.modeling_utils import ModelMixin
+from diffusers.utils import BaseOutput
+
+
+IMF_PRESET_CONFIGS: Dict[str, Dict[str, object]] = {
+    "iMF-B/2": {
+        "sample_size": 32,
+        "patch_size": 2,
+        "hidden_size": 768,
+        "depth": 12,
+        "num_attention_heads": 12,
+        "aux_head_depth": 8,
+    },
+    "iMF-M/2": {
+        "sample_size": 32,
+        "patch_size": 2,
+        "hidden_size": 768,
+        "depth": 24,
+        "num_attention_heads": 12,
+        "aux_head_depth": 8,
+    },
+    "iMF-L/2": {
+        "sample_size": 32,
+        "patch_size": 2,
+        "hidden_size": 1024,
+        "depth": 32,
+        "num_attention_heads": 16,
+        "aux_head_depth": 8,
+    },
+    "iMF-XL/2": {
+        "sample_size": 32,
+        "patch_size": 2,
+        "hidden_size": 1024,
+        "depth": 48,
+        "num_attention_heads": 16,
+        "aux_head_depth": 8,
+    },
+}
+
+
+@dataclass
+class IMFTransformer2DModelOutput(BaseOutput):
+    velocity_u: torch.Tensor
+    velocity_v: Optional[torch.Tensor] = None
+
+
+def remap_legacy_state_dict(state_dict: Dict[str, torch.Tensor]) -> Dict[str, torch.Tensor]:
+    """Map legacy iMF / Flax-style keys to native IMFTransformer2DModel keys."""
+    remapped: Dict[str, torch.Tensor] = {}
+    for key, value in state_dict.items():
+        new_key = key
+        for prefix in ("net.", "transformer."):
+            if new_key.startswith(prefix):
+                new_key = new_key[len(prefix) :]
+                break
+        new_key = new_key.replace(".kernel", ".weight")
+        new_key = new_key.replace("._flax_linear.", ".linear.")
+        new_key = new_key.replace("._flax_embedding.", ".embedding.")
+        remapped[new_key] = value
+    return remapped
+
+
+def config_from_legacy(config: Dict[str, object]) -> Dict[str, object]:
+    model_type = config.get("model_type") or config.get("model_str") or config.get("model_name")
+    if model_type not in IMF_PRESET_CONFIGS:
+        raise ValueError(f"Unknown iMF preset '{model_type}'. Known: {list(IMF_PRESET_CONFIGS)}")
+
+    preset = dict(IMF_PRESET_CONFIGS[model_type])
+    preset["num_classes"] = int(config.get("num_class_embeds") or config.get("num_classes") or 1000)
+    preset["model_type"] = model_type
+    if config.get("sample_size") is not None:
+        preset["sample_size"] = int(config["sample_size"])
+    if config.get("in_channels") is not None:
+        preset["in_channels"] = int(config["in_channels"])
+    return preset
+
+
+def _unsqueeze(t: torch.Tensor, dim: int) -> torch.Tensor:
+    return t.unsqueeze(dim)
+
+
+class _ScaledLinear(nn.Module):
+    def __init__(
+        self,
+        in_features: int,
+        out_features: int,
+        bias: bool = True,
+        weight_init: str = "scaled_variance",
+        init_constant: float = 1.0,
+        bias_init: str = "zeros",
+    ):
+        super().__init__()
+        self.linear = nn.Linear(in_features, out_features, bias=bias)
+        if weight_init == "scaled_variance":
+            std = init_constant / sqrt(in_features)
+            nn.init.normal_(self.linear.weight, std=std)
+        elif weight_init == "zeros":
+            nn.init.zeros_(self.linear.weight)
+        else:
+            raise ValueError(f"Invalid weight_init: {weight_init}")
+        if bias:
+            if bias_init == "zeros":
+                nn.init.zeros_(self.linear.bias)
+            else:
+                raise ValueError(f"Invalid bias_init: {bias_init}")
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        return self.linear(x)
+
+
+class _ScaledEmbedding(nn.Module):
+    def __init__(
+        self,
+        num_embeddings: int,
+        embedding_dim: int,
+        weight_init: str = "scaled_variance",
+        init_constant: float = 1.0,
+    ):
+        super().__init__()
+        self.embedding = nn.Embedding(num_embeddings, embedding_dim)
+        if weight_init == "scaled_variance":
+            std = init_constant / sqrt(embedding_dim)
+            nn.init.normal_(self.embedding.weight, std=std)
+        else:
+            raise ValueError(f"Invalid weight_init: {weight_init}")
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        return self.embedding(x)
+
+
+class _RMSNorm(nn.Module):
+    def __init__(self, dim: int, eps: float = 1e-6):
+        super().__init__()
+        self.weight = nn.Parameter(torch.ones(dim))
+        self.eps = eps
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        mean_square = torch.mean(torch.square(x), dim=-1, keepdim=True)
+        output = x * torch.rsqrt(mean_square + self.eps)
+        return output.to(x.dtype) * self.weight
+
+
+class _SwiGLUMlp(nn.Module):
+    def __init__(
+        self,
+        in_features: int,
+        hidden_features: int,
+        weight_init: str = "scaled_variance",
+        weight_init_constant: float = 1.0,
+    ):
+        super().__init__()
+        init_kwargs = dict(bias=False, weight_init=weight_init, init_constant=weight_init_constant)
+        self.w1 = _ScaledLinear(in_features, hidden_features, **init_kwargs)
+        self.w3 = _ScaledLinear(in_features, hidden_features, **init_kwargs)
+        self.w2 = _ScaledLinear(hidden_features, in_features, **init_kwargs)
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        return self.w2(F.silu(self.w1(x)) * self.w3(x))
+
+
+class IMFTimestepEmbedder(nn.Module):
+    def __init__(
+        self,
+        hidden_size: int,
+        frequency_embedding_size: int = 256,
+        init_constant: float = 1.0,
+    ):
+        super().__init__()
+        self.mlp = nn.Sequential(
+            _ScaledLinear(frequency_embedding_size, hidden_size, bias=True, init_constant=init_constant),
+            nn.SiLU(),
+            _ScaledLinear(hidden_size, hidden_size, bias=True, init_constant=init_constant),
+        )
+        self.frequency_embedding_size = frequency_embedding_size
+
+    @staticmethod
+    def timestep_embedding(t: torch.Tensor, dim: int, max_period: int = 10000) -> torch.Tensor:
+        half = dim // 2
+        freqs = torch.exp(
+            -math.log(max_period) * torch.arange(start=0, end=half, dtype=torch.float32, device=t.device) / half
+        )
+        args = t[:, None].float() * freqs[None]
+        embedding = torch.cat([torch.cos(args), torch.sin(args)], dim=-1)
+        if dim % 2:
+            embedding = torch.cat([embedding, torch.zeros_like(embedding[:, :1])], dim=-1)
+        return embedding
+
+    def forward(self, t: torch.Tensor) -> torch.Tensor:
+        embedding = self.timestep_embedding(t, self.frequency_embedding_size)
+        return self.mlp(embedding.to(dtype=next(self.mlp.parameters()).dtype))
+
+
+class IMFLabelEmbedder(nn.Module):
+    def __init__(self, num_classes: int, hidden_size: int, init_constant: float = 1.0):
+        super().__init__()
+        self.embedding_table = _ScaledEmbedding(num_classes + 1, hidden_size, init_constant=init_constant)
+
+    def forward(self, labels: torch.Tensor) -> torch.Tensor:
+        return self.embedding_table(labels)
+
+
+class IMFPatchEmbedder(nn.Module):
+    def __init__(self, input_size: int, patch_size: int, in_channels: int, hidden_size: int, bias: bool = True):
+        super().__init__()
+        self.patch_size = (patch_size, patch_size)
+        self.num_patches = (input_size // patch_size) ** 2
+        self.proj = nn.Conv2d(
+            in_channels,
+            hidden_size,
+            kernel_size=patch_size,
+            stride=patch_size,
+            bias=bias,
+        )
+        kh = kw = patch_size
+        fan_in = kh * kw * in_channels
+        fan_out = hidden_size
+        limit = math.sqrt(6.0 / (fan_in + fan_out))
+        nn.init.uniform_(self.proj.weight, -limit, limit)
+        if bias:
+            nn.init.zeros_(self.proj.bias)
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        x = self.proj(x)
+        return x.flatten(2).transpose(1, 2)
+
+
+def precompute_rope_freqs(dim: int, seq_len: int, theta: float = 10000.0, device: torch.device | None = None):
+    freqs = 1.0 / (theta ** (torch.arange(0, dim, 2, dtype=torch.float32, device=device) / dim))
+    positions = torch.arange(seq_len, dtype=torch.float32, device=device)
+    freqs_cis = torch.outer(positions, freqs)
+    return torch.complex(torch.cos(freqs_cis), torch.sin(freqs_cis))
+
+
+def apply_rotary_pos_emb(x: torch.Tensor, freqs_cis: torch.Tensor) -> torch.Tensor:
+    x_ = x.float().reshape(*x.shape[:-1], -1, 2)
+    x_complex = torch.view_as_complex(x_)
+    freqs_cis = freqs_cis.unsqueeze(0).unsqueeze(2).to(x.device)
+    x_rotated = torch.view_as_real(x_complex * freqs_cis).flatten(-2)
+    return x_rotated.to(dtype=x.dtype)
+
+
+class IMFRoPEAttention(nn.Module):
+    def __init__(self, hidden_size: int, num_heads: int, weight_init_constant: float = 0.32):
+        super().__init__()
+        self.hidden_size = hidden_size
+        self.num_heads = num_heads
+        self.head_dim = hidden_size // num_heads
+        init_kwargs = dict(bias=False, init_constant=weight_init_constant)
+        self.q_proj = _ScaledLinear(hidden_size, hidden_size, **init_kwargs)
+        self.k_proj = _ScaledLinear(hidden_size, hidden_size, **init_kwargs)
+        self.v_proj = _ScaledLinear(hidden_size, hidden_size, **init_kwargs)
+        self.out_proj = _ScaledLinear(hidden_size, hidden_size, **init_kwargs)
+        self.q_norm = _RMSNorm(self.head_dim)
+        self.k_norm = _RMSNorm(self.head_dim)
+
+    def forward(self, x: torch.Tensor, rope_freqs: torch.Tensor) -> torch.Tensor:
+        batch, seq_len, _ = x.shape
+        q = self.q_proj(x).reshape(batch, seq_len, self.num_heads, self.head_dim)
+        k = self.k_proj(x).reshape(batch, seq_len, self.num_heads, self.head_dim)
+        v = self.v_proj(x).reshape(batch, seq_len, self.num_heads, self.head_dim)
+        q = apply_rotary_pos_emb(self.q_norm(q), rope_freqs)
+        k = apply_rotary_pos_emb(self.k_norm(k), rope_freqs)
+        query = q / math.sqrt(self.head_dim)
+        attn_weights = torch.einsum("bqhd,bkhd->bhqk", query, k)
+        attn_weights = F.softmax(attn_weights, dim=-1, dtype=torch.float32).to(v.dtype)
+        attn = torch.einsum("bhqk,bkhd->bqhd", attn_weights, v)
+        return self.out_proj(attn.reshape(batch, seq_len, self.hidden_size))
+
+
+class IMFTransformerBlock(nn.Module):
+    def __init__(self, hidden_size: int, num_heads: int, mlp_ratio: float, weight_init_constant: float):
+        super().__init__()
+        self.norm1 = _RMSNorm(hidden_size)
+        self.attn = IMFRoPEAttention(hidden_size, num_heads, weight_init_constant=weight_init_constant)
+        self.norm2 = _RMSNorm(hidden_size)
+        mlp_hidden_dim = int(hidden_size * mlp_ratio)
+        self.mlp = _SwiGLUMlp(hidden_size, mlp_hidden_dim, weight_init_constant=weight_init_constant)
+        self.attn_scale = nn.Parameter(torch.zeros(hidden_size))
+        self.mlp_scale = nn.Parameter(torch.zeros(hidden_size))
+
+    def forward(self, x: torch.Tensor, rope_freqs: torch.Tensor) -> torch.Tensor:
+        x = x + self.attn(self.norm1(x), rope_freqs) * self.attn_scale
+        x = x + self.mlp(self.norm2(x)) * self.mlp_scale
+        return x
+
+
+class IMFFinalLayer(nn.Module):
+    def __init__(self, hidden_size: int, patch_size: int, out_channels: int):
+        super().__init__()
+        self.norm = _RMSNorm(hidden_size)
+        self.linear = _ScaledLinear(
+            hidden_size,
+            patch_size * patch_size * out_channels,
+            bias=True,
+            weight_init="zeros",
+            init_constant=1.0,
+        )
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        return self.linear(self.norm(x))
+
+
+class IMFTransformer2DModel(ModelMixin, ConfigMixin):
+    _supports_gradient_checkpointing = True
+
+    @register_to_config
+    def __init__(
+        self,
+        sample_size: int = 32,
+        patch_size: int = 2,
+        in_channels: int = 4,
+        hidden_size: int = 768,
+        depth: int = 12,
+        num_attention_heads: int = 12,
+        mlp_ratio: float = 8 / 3,
+        num_classes: int = 1000,
+        aux_head_depth: int = 8,
+        num_class_tokens: int = 8,
+        num_time_tokens: int = 4,
+        num_cfg_tokens: int = 4,
+        num_interval_tokens: int = 2,
+        token_init_constant: float = 1.0,
+        embedding_init_constant: float = 1.0,
+        weight_init_constant: float = 0.32,
+        eval_mode: bool = True,
+        model_type: str | None = None,
+        num_class_embeds: int | None = None,
+    ):
+        super().__init__()
+        if num_class_embeds is not None:
+            num_classes = int(num_class_embeds)
+        if model_type in IMF_PRESET_CONFIGS:
+            preset = IMF_PRESET_CONFIGS[model_type]
+            sample_size = int(preset["sample_size"])
+            patch_size = int(preset["patch_size"])
+            hidden_size = int(preset["hidden_size"])
+            depth = int(preset["depth"])
+            num_attention_heads = int(preset["num_attention_heads"])
+            aux_head_depth = int(preset["aux_head_depth"])
+
+        self.sample_size = sample_size
+        self.patch_size = patch_size
+        self.in_channels = in_channels
+        self.out_channels = in_channels
+        self.hidden_size = hidden_size
+        self.depth = depth
+        self.num_attention_heads = num_attention_heads
+        self.num_classes = num_classes
+        self.aux_head_depth = aux_head_depth
+        self.num_class_tokens = num_class_tokens
+        self.num_time_tokens = num_time_tokens
+        self.num_cfg_tokens = num_cfg_tokens
+        self.num_interval_tokens = num_interval_tokens
+        self.eval_mode = eval_mode
+        self.gradient_checkpointing = False
+
+        self.x_embedder = IMFPatchEmbedder(sample_size, patch_size, in_channels, hidden_size, bias=True)
+        embed_kwargs = dict(hidden_size=hidden_size, init_constant=embedding_init_constant)
+        self.h_embedder = IMFTimestepEmbedder(**embed_kwargs)
+        self.omega_embedder = IMFTimestepEmbedder(**embed_kwargs)
+        self.cfg_t_start_embedder = IMFTimestepEmbedder(**embed_kwargs)
+        self.cfg_t_end_embedder = IMFTimestepEmbedder(**embed_kwargs)
+        self.y_embedder = IMFLabelEmbedder(num_classes, hidden_size, init_constant=embedding_init_constant)
+
+        token_std = token_init_constant / math.sqrt(hidden_size)
+        self.time_tokens = nn.Parameter(torch.randn(num_time_tokens, hidden_size) * token_std)
+        self.class_tokens = nn.Parameter(torch.randn(num_class_tokens, hidden_size) * token_std)
+        self.omega_tokens = nn.Parameter(torch.randn(num_cfg_tokens, hidden_size) * token_std)
+        self.t_min_tokens = nn.Parameter(torch.randn(num_interval_tokens, hidden_size) * token_std)
+        self.t_max_tokens = nn.Parameter(torch.randn(num_interval_tokens, hidden_size) * token_std)
+
+        total_tokens = (
+            self.x_embedder.num_patches
+            + num_class_tokens
+            + num_cfg_tokens
+            + 2 * num_interval_tokens
+            + num_time_tokens
+        )
+        self.prefix_tokens = num_class_tokens + num_cfg_tokens + 2 * num_interval_tokens + num_time_tokens
+        self.head_dim = hidden_size // num_attention_heads
+        self.register_buffer(
+            "rope_freqs",
+            precompute_rope_freqs(self.head_dim, total_tokens),
+            persistent=False,
+        )
+
+        shared_depth = depth - aux_head_depth
+        block_kwargs = dict(
+            hidden_size=hidden_size,
+            num_heads=num_attention_heads,
+            mlp_ratio=mlp_ratio,
+            weight_init_constant=weight_init_constant,
+        )
+        self.shared_blocks = nn.ModuleList([IMFTransformerBlock(**block_kwargs) for _ in range(shared_depth)])
+        self.u_heads = nn.ModuleList([IMFTransformerBlock(**block_kwargs) for _ in range(aux_head_depth)])
+        self.v_heads = nn.ModuleList(
+            [IMFTransformerBlock(**block_kwargs) for _ in range(aux_head_depth if not eval_mode else 0)]
+        )
+        self.u_final_layer = IMFFinalLayer(hidden_size, patch_size, in_channels)
+        self.v_final_layer = IMFFinalLayer(hidden_size, patch_size, in_channels)
+
+    def unpatchify(self, x: torch.Tensor) -> torch.Tensor:
+        c = self.out_channels
+        p = self.patch_size
+        h = w = int(x.shape[1] ** 0.5)
+        x = x.reshape(x.shape[0], h, w, p, p, c)
+        x = torch.einsum("nhwpqc->nchpwq", x)
+        return x.reshape(x.shape[0], c, h * p, w * p)
+
+    def _build_sequence(
+        self,
+        sample: torch.Tensor,
+        time_gap: torch.Tensor,
+        guidance_scale: torch.Tensor,
+        guidance_interval_start: torch.Tensor,
+        guidance_interval_end: torch.Tensor,
+        class_labels: torch.Tensor,
+    ) -> torch.Tensor:
+        x_embed = self.x_embedder(sample)
+        h_embed = self.h_embedder(time_gap)
+        omega_embed = self.omega_embedder(1 - 1 / guidance_scale)
+        t_min_embed = self.cfg_t_start_embedder(guidance_interval_start)
+        t_max_embed = self.cfg_t_end_embedder(guidance_interval_end)
+        y_embed = self.y_embedder(class_labels)
+
+        time_tokens = self.time_tokens + _unsqueeze(h_embed, 1)
+        omega_tokens = self.omega_tokens + _unsqueeze(omega_embed, 1)
+        t_min_tokens = self.t_min_tokens + _unsqueeze(t_min_embed, 1)
+        t_max_tokens = self.t_max_tokens + _unsqueeze(t_max_embed, 1)
+        class_tokens = self.class_tokens + _unsqueeze(y_embed, 1)
+
+        return torch.cat(
+            [class_tokens, omega_tokens, t_min_tokens, t_max_tokens, time_tokens, x_embed],
+            dim=1,
+        )
+
+    def forward(
+        self,
+        sample: torch.Tensor,
+        timestep: torch.Tensor,
+        class_labels: torch.Tensor,
+        time_gap: Optional[torch.Tensor] = None,
+        guidance_scale: Optional[torch.Tensor] = None,
+        guidance_interval_start: Optional[torch.Tensor] = None,
+        guidance_interval_end: Optional[torch.Tensor] = None,
+        return_dict: bool = True,
+    ) -> Union[IMFTransformer2DModelOutput, Tuple[torch.Tensor, ...]]:
+        batch_size = sample.shape[0]
+        timestep = self._expand_batch(timestep, batch_size, sample.device, sample.dtype)
+        if time_gap is None:
+            time_gap = timestep
+        else:
+            time_gap = self._expand_batch(time_gap, batch_size, sample.device, sample.dtype)
+        if guidance_scale is None:
+            guidance_scale = torch.ones(batch_size, device=sample.device, dtype=sample.dtype)
+        else:
+            guidance_scale = self._expand_batch(guidance_scale, batch_size, sample.device, sample.dtype)
+        if guidance_interval_start is None:
+            guidance_interval_start = torch.zeros(batch_size, device=sample.device, dtype=sample.dtype)
+        else:
+            guidance_interval_start = self._expand_batch(
+                guidance_interval_start, batch_size, sample.device, sample.dtype
+            )
+        if guidance_interval_end is None:
+            guidance_interval_end = torch.ones(batch_size, device=sample.device, dtype=sample.dtype)
+        else:
+            guidance_interval_end = self._expand_batch(
+                guidance_interval_end, batch_size, sample.device, sample.dtype
+            )
+
+        class_labels = class_labels.reshape(-1).long()
+        seq = self._build_sequence(
+            sample,
+            time_gap,
+            guidance_scale,
+            guidance_interval_start,
+            guidance_interval_end,
+            class_labels,
+        )
+
+        for block in self.shared_blocks:
+            if self.training and self.gradient_checkpointing:
+                seq = torch.utils.checkpoint.checkpoint(block, seq, self.rope_freqs, use_reentrant=False)
+            else:
+                seq = block(seq, self.rope_freqs)
+
+        u_seq = v_seq = seq
+        for block in self.u_heads:
+            u_seq = block(u_seq, self.rope_freqs)
+        for block in self.v_heads:
+            v_seq = block(v_seq, self.rope_freqs)
+
+        u_tokens = u_seq[:, self.prefix_tokens :]
+        velocity_u = self.unpatchify(self.u_final_layer(u_tokens))
+
+        velocity_v = None
+        if len(self.v_heads) > 0:
+            v_tokens = v_seq[:, self.prefix_tokens :]
+            velocity_v = self.unpatchify(self.v_final_layer(v_tokens))
+
+        if not return_dict:
+            if velocity_v is None:
+                return (velocity_u,)
+            return (velocity_u, velocity_v)
+        return IMFTransformer2DModelOutput(velocity_u=velocity_u, velocity_v=velocity_v)
+
+    @staticmethod
+    def _expand_batch(
+        value: torch.Tensor,
+        batch_size: int,
+        device: torch.device,
+        dtype: torch.dtype,
+    ) -> torch.Tensor:
+        value = torch.as_tensor(value, device=device, dtype=dtype)
+        if value.ndim == 0:
+            value = value.reshape(1).repeat(batch_size)
+        else:
+            value = value.reshape(-1)
+            if value.shape[0] == 1 and batch_size > 1:
+                value = value.repeat(batch_size)
+        return value
+
+    @classmethod
+    def from_imf_checkpoint(
+        cls,
+        checkpoint_path: str,
+        model_type: str | None = None,
+        map_location: str = "cpu",
+        strict: bool = True,
+        eval_mode: bool = True,
+    ) -> Tuple["IMFTransformer2DModel", Dict[str, object]]:
+        checkpoint = torch.load(checkpoint_path, map_location=map_location, weights_only=False)
+        if isinstance(checkpoint, Mapping) and "state_dict" in checkpoint:
+            state_dict = checkpoint["state_dict"]
+        elif isinstance(checkpoint, Mapping) and any(k.startswith("net.") for k in checkpoint):
+            state_dict = checkpoint
+        else:
+            state_dict = checkpoint
+
+        inferred_type = model_type
+        if inferred_type is None and isinstance(checkpoint, Mapping):
+            args = checkpoint.get("args")
+            if isinstance(args, argparse.Namespace):
+                inferred_type = getattr(args, "model", None) or getattr(args, "model_str", None)
+            elif isinstance(args, Mapping):
+                inferred_type = args.get("model") or args.get("model_str")
+
+        if inferred_type is None:
+            raise ValueError("model_type must be provided when loading a raw state dict checkpoint.")
+
+        config = dict(IMF_PRESET_CONFIGS[inferred_type])
+        config["model_type"] = inferred_type
+        config["eval_mode"] = eval_mode
+        model = cls(**config)
+        model.load_state_dict(remap_legacy_state_dict(state_dict), strict=strict)
+        metadata = {"checkpoint_path": checkpoint_path, "model_type": inferred_type}
+        return model, metadata
+
+    def to_imf_checkpoint(self, prefix: str = "net.") -> Dict[str, torch.Tensor]:
+        state: Dict[str, torch.Tensor] = {}
+        for key, value in self.state_dict().items():
+            if key == "rope_freqs":
+                continue
+            state[f"{prefix}{key}"] = value.detach().cpu()
+        return state
+
+    @property
+    def net(self):
+        return self
+
+
+IMFDiffusersModel = IMFTransformer2DModel

iMF-L-2/vae/config.json

@@ -0,0 +1,38 @@
+{
+  "_class_name": "AutoencoderKL",
+  "_diffusers_version": "0.36.0",
+  "_name_or_path": "stabilityai/sd-vae-ft-mse",
+  "act_fn": "silu",
+  "block_out_channels": [
+    128,
+    256,
+    512,
+    512
+  ],
+  "down_block_types": [
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D"
+  ],
+  "force_upcast": true,
+  "in_channels": 3,
+  "latent_channels": 4,
+  "latents_mean": null,
+  "latents_std": null,
+  "layers_per_block": 2,
+  "mid_block_add_attention": true,
+  "norm_num_groups": 32,
+  "out_channels": 3,
+  "sample_size": 256,
+  "scaling_factor": 0.18215,
+  "shift_factor": null,
+  "up_block_types": [
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D"
+  ],
+  "use_post_quant_conv": true,
+  "use_quant_conv": true
+}

iMF-L-2/vae/diffusion_pytorch_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a1d993488569e928462932c8c38a0760b874d166399b14414135bd9c42df5815
+size 334643276

iMF-XL-2/README.md

@@ -0,0 +1,54 @@
+---
+license: mit
+library_name: diffusers
+pipeline_tag: text-to-image
+tags:
+- diffusers
+- imf
+- image-generation
+- class-conditional
+inference: true
+widget:
+- output:
+    url: demo.png
+language:
+- en
+---
+
+# iMF-XL-2
+
+Self-contained Diffusers variant for **iMF-XL/2** (Improved Mean Flows).
+
+## Demo
+
+![iMF-XL-2 demo](demo.png)
+
+Class-conditional sample (ImageNet class **207**, golden retriever), 1 step, CFG 1.8, interval [0.0, 1.0], seed 42.
+
+## Load
+
+```python
+from pathlib import Path
+from diffusers import DiffusionPipeline
+import torch
+
+model_dir = Path("iMF-XL-2")
+pipe = DiffusionPipeline.from_pretrained(
+    str(model_dir),
+    local_files_only=True,
+    custom_pipeline=str(model_dir / "pipeline.py"),
+    trust_remote_code=True,
+    torch_dtype=torch.bfloat16,
+).to("cuda")
+
+generator = torch.Generator(device="cuda").manual_seed(42)
+image = pipe(
+    class_labels="golden retriever",
+    num_inference_steps=1,
+    guidance_scale=1.8,
+    guidance_interval_start=0.0,
+    guidance_interval_end=1.0,
+    generator=generator,
+).images[0]
+image.save("demo.png")
+```

iMF-XL-2/model_index.json

@@ -0,0 +1,1021 @@
+{
+  "_class_name": [
+    "pipeline",
+    "IMFPipeline"
+  ],
+  "_diffusers_version": "0.36.0",
+  "scheduler": [
+    "diffusers",
+    "FlowMatchEulerDiscreteScheduler"
+  ],
+  "transformer": [
+    "transformer_imf",
+    "IMFTransformer2DModel"
+  ],
+  "vae": [
+    "diffusers",
+    "AutoencoderKL"
+  ],
+  "id2label": {
+    "0": "tench, Tinca tinca",
+    "1": "goldfish, Carassius auratus",
+    "2": "great white shark, white shark, man-eater, man-eating shark, Carcharodon carcharias",
+    "3": "tiger shark, Galeocerdo cuvieri",
+    "4": "hammerhead, hammerhead shark",
+    "5": "electric ray, crampfish, numbfish, torpedo",
+    "6": "stingray",
+    "7": "cock",
+    "8": "hen",
+    "9": "ostrich, Struthio camelus",
+    "10": "brambling, Fringilla montifringilla",
+    "11": "goldfinch, Carduelis carduelis",
+    "12": "house finch, linnet, Carpodacus mexicanus",
+    "13": "junco, snowbird",
+    "14": "indigo bunting, indigo finch, indigo bird, Passerina cyanea",
+    "15": "robin, American robin, Turdus migratorius",
+    "16": "bulbul",
+    "17": "jay",
+    "18": "magpie",
+    "19": "chickadee",
+    "20": "water ouzel, dipper",
+    "21": "kite",
+    "22": "bald eagle, American eagle, Haliaeetus leucocephalus",
+    "23": "vulture",
+    "24": "great grey owl, great gray owl, Strix nebulosa",
+    "25": "European fire salamander, Salamandra salamandra",
+    "26": "common newt, Triturus vulgaris",
+    "27": "eft",
+    "28": "spotted salamander, Ambystoma maculatum",
+    "29": "axolotl, mud puppy, Ambystoma mexicanum",
+    "30": "bullfrog, Rana catesbeiana",
+    "31": "tree frog, tree-frog",
+    "32": "tailed frog, bell toad, ribbed toad, tailed toad, Ascaphus trui",
+    "33": "loggerhead, loggerhead turtle, Caretta caretta",
+    "34": "leatherback turtle, leatherback, leathery turtle, Dermochelys coriacea",
+    "35": "mud turtle",
+    "36": "terrapin",
+    "37": "box turtle, box tortoise",
+    "38": "banded gecko",
+    "39": "common iguana, iguana, Iguana iguana",
+    "40": "American chameleon, anole, Anolis carolinensis",
+    "41": "whiptail, whiptail lizard",
+    "42": "agama",
+    "43": "frilled lizard, Chlamydosaurus kingi",
+    "44": "alligator lizard",
+    "45": "Gila monster, Heloderma suspectum",
+    "46": "green lizard, Lacerta viridis",
+    "47": "African chameleon, Chamaeleo chamaeleon",
+    "48": "Komodo dragon, Komodo lizard, dragon lizard, giant lizard, Varanus komodoensis",
+    "49": "African crocodile, Nile crocodile, Crocodylus niloticus",
+    "50": "American alligator, Alligator mississipiensis",
+    "51": "triceratops",
+    "52": "thunder snake, worm snake, Carphophis amoenus",
+    "53": "ringneck snake, ring-necked snake, ring snake",
+    "54": "hognose snake, puff adder, sand viper",
+    "55": "green snake, grass snake",
+    "56": "king snake, kingsnake",
+    "57": "garter snake, grass snake",
+    "58": "water snake",
+    "59": "vine snake",
+    "60": "night snake, Hypsiglena torquata",
+    "61": "boa constrictor, Constrictor constrictor",
+    "62": "rock python, rock snake, Python sebae",
+    "63": "Indian cobra, Naja naja",
+    "64": "green mamba",
+    "65": "sea snake",
+    "66": "horned viper, cerastes, sand viper, horned asp, Cerastes cornutus",
+    "67": "diamondback, diamondback rattlesnake, Crotalus adamanteus",
+    "68": "sidewinder, horned rattlesnake, Crotalus cerastes",
+    "69": "trilobite",
+    "70": "harvestman, daddy longlegs, Phalangium opilio",
+    "71": "scorpion",
+    "72": "black and gold garden spider, Argiope aurantia",
+    "73": "barn spider, Araneus cavaticus",
+    "74": "garden spider, Aranea diademata",
+    "75": "black widow, Latrodectus mactans",
+    "76": "tarantula",
+    "77": "wolf spider, hunting spider",
+    "78": "tick",
+    "79": "centipede",
+    "80": "black grouse",
+    "81": "ptarmigan",
+    "82": "ruffed grouse, partridge, Bonasa umbellus",
+    "83": "prairie chicken, prairie grouse, prairie fowl",
+    "84": "peacock",
+    "85": "quail",
+    "86": "partridge",
+    "87": "African grey, African gray, Psittacus erithacus",
+    "88": "macaw",
+    "89": "sulphur-crested cockatoo, Kakatoe galerita, Cacatua galerita",
+    "90": "lorikeet",
+    "91": "coucal",
+    "92": "bee eater",
+    "93": "hornbill",
+    "94": "hummingbird",
+    "95": "jacamar",
+    "96": "toucan",
+    "97": "drake",
+    "98": "red-breasted merganser, Mergus serrator",
+    "99": "goose",
+    "100": "black swan, Cygnus atratus",
+    "101": "tusker",
+    "102": "echidna, spiny anteater, anteater",
+    "103": "platypus, duckbill, duckbilled platypus, duck-billed platypus, Ornithorhynchus anatinus",
+    "104": "wallaby, brush kangaroo",
+    "105": "koala, koala bear, kangaroo bear, native bear, Phascolarctos cinereus",
+    "106": "wombat",
+    "107": "jellyfish",
+    "108": "sea anemone, anemone",
+    "109": "brain coral",
+    "110": "flatworm, platyhelminth",
+    "111": "nematode, nematode worm, roundworm",
+    "112": "conch",
+    "113": "snail",
+    "114": "slug",
+    "115": "sea slug, nudibranch",
+    "116": "chiton, coat-of-mail shell, sea cradle, polyplacophore",
+    "117": "chambered nautilus, pearly nautilus, nautilus",
+    "118": "Dungeness crab, Cancer magister",
+    "119": "rock crab, Cancer irroratus",
+    "120": "fiddler crab",
+    "121": "king crab, Alaska crab, Alaskan king crab, Alaska king crab, Paralithodes camtschatica",
+    "122": "American lobster, Northern lobster, Maine lobster, Homarus americanus",
+    "123": "spiny lobster, langouste, rock lobster, crawfish, crayfish, sea crawfish",
+    "124": "crayfish, crawfish, crawdad, crawdaddy",
+    "125": "hermit crab",
+    "126": "isopod",
+    "127": "white stork, Ciconia ciconia",
+    "128": "black stork, Ciconia nigra",
+    "129": "spoonbill",
+    "130": "flamingo",
+    "131": "little blue heron, Egretta caerulea",
+    "132": "American egret, great white heron, Egretta albus",
+    "133": "bittern",
+    "134": "crane",
+    "135": "limpkin, Aramus pictus",
+    "136": "European gallinule, Porphyrio porphyrio",
+    "137": "American coot, marsh hen, mud hen, water hen, Fulica americana",
+    "138": "bustard",
+    "139": "ruddy turnstone, Arenaria interpres",
+    "140": "red-backed sandpiper, dunlin, Erolia alpina",
+    "141": "redshank, Tringa totanus",
+    "142": "dowitcher",
+    "143": "oystercatcher, oyster catcher",
+    "144": "pelican",
+    "145": "king penguin, Aptenodytes patagonica",
+    "146": "albatross, mollymawk",
+    "147": "grey whale, gray whale, devilfish, Eschrichtius gibbosus, Eschrichtius robustus",
+    "148": "killer whale, killer, orca, grampus, sea wolf, Orcinus orca",
+    "149": "dugong, Dugong dugon",
+    "150": "sea lion",
+    "151": "Chihuahua",
+    "152": "Japanese spaniel",
+    "153": "Maltese dog, Maltese terrier, Maltese",
+    "154": "Pekinese, Pekingese, Peke",
+    "155": "Shih-Tzu",
+    "156": "Blenheim spaniel",
+    "157": "papillon",
+    "158": "toy terrier",
+    "159": "Rhodesian ridgeback",
+    "160": "Afghan hound, Afghan",
+    "161": "basset, basset hound",
+    "162": "beagle",
+    "163": "bloodhound, sleuthhound",
+    "164": "bluetick",
+    "165": "black-and-tan coonhound",
+    "166": "Walker hound, Walker foxhound",
+    "167": "English foxhound",
+    "168": "redbone",
+    "169": "borzoi, Russian wolfhound",
+    "170": "Irish wolfhound",
+    "171": "Italian greyhound",
+    "172": "whippet",
+    "173": "Ibizan hound, Ibizan Podenco",
+    "174": "Norwegian elkhound, elkhound",
+    "175": "otterhound, otter hound",
+    "176": "Saluki, gazelle hound",
+    "177": "Scottish deerhound, deerhound",
+    "178": "Weimaraner",
+    "179": "Staffordshire bullterrier, Staffordshire bull terrier",
+    "180": "American Staffordshire terrier, Staffordshire terrier, American pit bull terrier, pit bull terrier",
+    "181": "Bedlington terrier",
+    "182": "Border terrier",
+    "183": "Kerry blue terrier",
+    "184": "Irish terrier",
+    "185": "Norfolk terrier",
+    "186": "Norwich terrier",
+    "187": "Yorkshire terrier",
+    "188": "wire-haired fox terrier",
+    "189": "Lakeland terrier",
+    "190": "Sealyham terrier, Sealyham",
+    "191": "Airedale, Airedale terrier",
+    "192": "cairn, cairn terrier",
+    "193": "Australian terrier",
+    "194": "Dandie Dinmont, Dandie Dinmont terrier",
+    "195": "Boston bull, Boston terrier",
+    "196": "miniature schnauzer",
+    "197": "giant schnauzer",
+    "198": "standard schnauzer",
+    "199": "Scotch terrier, Scottish terrier, Scottie",
+    "200": "Tibetan terrier, chrysanthemum dog",
+    "201": "silky terrier, Sydney silky",
+    "202": "soft-coated wheaten terrier",
+    "203": "West Highland white terrier",
+    "204": "Lhasa, Lhasa apso",
+    "205": "flat-coated retriever",
+    "206": "curly-coated retriever",
+    "207": "golden retriever",
+    "208": "Labrador retriever",
+    "209": "Chesapeake Bay retriever",
+    "210": "German short-haired pointer",
+    "211": "vizsla, Hungarian pointer",
+    "212": "English setter",
+    "213": "Irish setter, red setter",
+    "214": "Gordon setter",
+    "215": "Brittany spaniel",
+    "216": "clumber, clumber spaniel",
+    "217": "English springer, English springer spaniel",
+    "218": "Welsh springer spaniel",
+    "219": "cocker spaniel, English cocker spaniel, cocker",
+    "220": "Sussex spaniel",
+    "221": "Irish water spaniel",
+    "222": "kuvasz",
+    "223": "schipperke",
+    "224": "groenendael",
+    "225": "malinois",
+    "226": "briard",
+    "227": "kelpie",
+    "228": "komondor",
+    "229": "Old English sheepdog, bobtail",
+    "230": "Shetland sheepdog, Shetland sheep dog, Shetland",
+    "231": "collie",
+    "232": "Border collie",
+    "233": "Bouvier des Flandres, Bouviers des Flandres",
+    "234": "Rottweiler",
+    "235": "German shepherd, German shepherd dog, German police dog, alsatian",
+    "236": "Doberman, Doberman pinscher",
+    "237": "miniature pinscher",
+    "238": "Greater Swiss Mountain dog",
+    "239": "Bernese mountain dog",
+    "240": "Appenzeller",
+    "241": "EntleBucher",
+    "242": "boxer",
+    "243": "bull mastiff",
+    "244": "Tibetan mastiff",
+    "245": "French bulldog",
+    "246": "Great Dane",
+    "247": "Saint Bernard, St Bernard",
+    "248": "Eskimo dog, husky",
+    "249": "malamute, malemute, Alaskan malamute",
+    "250": "Siberian husky",
+    "251": "dalmatian, coach dog, carriage dog",
+    "252": "affenpinscher, monkey pinscher, monkey dog",
+    "253": "basenji",
+    "254": "pug, pug-dog",
+    "255": "Leonberg",
+    "256": "Newfoundland, Newfoundland dog",
+    "257": "Great Pyrenees",
+    "258": "Samoyed, Samoyede",
+    "259": "Pomeranian",
+    "260": "chow, chow chow",
+    "261": "keeshond",
+    "262": "Brabancon griffon",
+    "263": "Pembroke, Pembroke Welsh corgi",
+    "264": "Cardigan, Cardigan Welsh corgi",
+    "265": "toy poodle",
+    "266": "miniature poodle",
+    "267": "standard poodle",
+    "268": "Mexican hairless",
+    "269": "timber wolf, grey wolf, gray wolf, Canis lupus",
+    "270": "white wolf, Arctic wolf, Canis lupus tundrarum",
+    "271": "red wolf, maned wolf, Canis rufus, Canis niger",
+    "272": "coyote, prairie wolf, brush wolf, Canis latrans",
+    "273": "dingo, warrigal, warragal, Canis dingo",
+    "274": "dhole, Cuon alpinus",
+    "275": "African hunting dog, hyena dog, Cape hunting dog, Lycaon pictus",
+    "276": "hyena, hyaena",
+    "277": "red fox, Vulpes vulpes",
+    "278": "kit fox, Vulpes macrotis",
+    "279": "Arctic fox, white fox, Alopex lagopus",
+    "280": "grey fox, gray fox, Urocyon cinereoargenteus",
+    "281": "tabby, tabby cat",
+    "282": "tiger cat",
+    "283": "Persian cat",
+    "284": "Siamese cat, Siamese",
+    "285": "Egyptian cat",
+    "286": "cougar, puma, catamount, mountain lion, painter, panther, Felis concolor",
+    "287": "lynx, catamount",
+    "288": "leopard, Panthera pardus",
+    "289": "snow leopard, ounce, Panthera uncia",
+    "290": "jaguar, panther, Panthera onca, Felis onca",
+    "291": "lion, king of beasts, Panthera leo",
+    "292": "tiger, Panthera tigris",
+    "293": "cheetah, chetah, Acinonyx jubatus",
+    "294": "brown bear, bruin, Ursus arctos",
+    "295": "American black bear, black bear, Ursus americanus, Euarctos americanus",
+    "296": "ice bear, polar bear, Ursus Maritimus, Thalarctos maritimus",
+    "297": "sloth bear, Melursus ursinus, Ursus ursinus",
+    "298": "mongoose",
+    "299": "meerkat, mierkat",
+    "300": "tiger beetle",
+    "301": "ladybug, ladybeetle, lady beetle, ladybird, ladybird beetle",
+    "302": "ground beetle, carabid beetle",
+    "303": "long-horned beetle, longicorn, longicorn beetle",
+    "304": "leaf beetle, chrysomelid",
+    "305": "dung beetle",
+    "306": "rhinoceros beetle",
+    "307": "weevil",
+    "308": "fly",
+    "309": "bee",
+    "310": "ant, emmet, pismire",
+    "311": "grasshopper, hopper",
+    "312": "cricket",
+    "313": "walking stick, walkingstick, stick insect",
+    "314": "cockroach, roach",
+    "315": "mantis, mantid",
+    "316": "cicada, cicala",
+    "317": "leafhopper",
+    "318": "lacewing, lacewing fly",
+    "319": "dragonfly, darning needle, devils darning needle, sewing needle, snake feeder, snake doctor, mosquito hawk, skeeter hawk",
+    "320": "damselfly",
+    "321": "admiral",
+    "322": "ringlet, ringlet butterfly",
+    "323": "monarch, monarch butterfly, milkweed butterfly, Danaus plexippus",
+    "324": "cabbage butterfly",
+    "325": "sulphur butterfly, sulfur butterfly",
+    "326": "lycaenid, lycaenid butterfly",
+    "327": "starfish, sea star",
+    "328": "sea urchin",
+    "329": "sea cucumber, holothurian",
+    "330": "wood rabbit, cottontail, cottontail rabbit",
+    "331": "hare",
+    "332": "Angora, Angora rabbit",
+    "333": "hamster",
+    "334": "porcupine, hedgehog",
+    "335": "fox squirrel, eastern fox squirrel, Sciurus niger",
+    "336": "marmot",
+    "337": "beaver",
+    "338": "guinea pig, Cavia cobaya",
+    "339": "sorrel",
+    "340": "zebra",
+    "341": "hog, pig, grunter, squealer, Sus scrofa",
+    "342": "wild boar, boar, Sus scrofa",
+    "343": "warthog",
+    "344": "hippopotamus, hippo, river horse, Hippopotamus amphibius",
+    "345": "ox",
+    "346": "water buffalo, water ox, Asiatic buffalo, Bubalus bubalis",
+    "347": "bison",
+    "348": "ram, tup",
+    "349": "bighorn, bighorn sheep, cimarron, Rocky Mountain bighorn, Rocky Mountain sheep, Ovis canadensis",
+    "350": "ibex, Capra ibex",
+    "351": "hartebeest",
+    "352": "impala, Aepyceros melampus",
+    "353": "gazelle",
+    "354": "Arabian camel, dromedary, Camelus dromedarius",
+    "355": "llama",
+    "356": "weasel",
+    "357": "mink",
+    "358": "polecat, fitch, foulmart, foumart, Mustela putorius",
+    "359": "black-footed ferret, ferret, Mustela nigripes",
+    "360": "otter",
+    "361": "skunk, polecat, wood pussy",
+    "362": "badger",
+    "363": "armadillo",
+    "364": "three-toed sloth, ai, Bradypus tridactylus",
+    "365": "orangutan, orang, orangutang, Pongo pygmaeus",
+    "366": "gorilla, Gorilla gorilla",
+    "367": "chimpanzee, chimp, Pan troglodytes",
+    "368": "gibbon, Hylobates lar",
+    "369": "siamang, Hylobates syndactylus, Symphalangus syndactylus",
+    "370": "guenon, guenon monkey",
+    "371": "patas, hussar monkey, Erythrocebus patas",
+    "372": "baboon",
+    "373": "macaque",
+    "374": "langur",
+    "375": "colobus, colobus monkey",
+    "376": "proboscis monkey, Nasalis larvatus",
+    "377": "marmoset",
+    "378": "capuchin, ringtail, Cebus capucinus",
+    "379": "howler monkey, howler",
+    "380": "titi, titi monkey",
+    "381": "spider monkey, Ateles geoffroyi",
+    "382": "squirrel monkey, Saimiri sciureus",
+    "383": "Madagascar cat, ring-tailed lemur, Lemur catta",
+    "384": "indri, indris, Indri indri, Indri brevicaudatus",
+    "385": "Indian elephant, Elephas maximus",
+    "386": "African elephant, Loxodonta africana",
+    "387": "lesser panda, red panda, panda, bear cat, cat bear, Ailurus fulgens",
+    "388": "giant panda, panda, panda bear, coon bear, Ailuropoda melanoleuca",
+    "389": "barracouta, snoek",
+    "390": "eel",
+    "391": "coho, cohoe, coho salmon, blue jack, silver salmon, Oncorhynchus kisutch",
+    "392": "rock beauty, Holocanthus tricolor",
+    "393": "anemone fish",
+    "394": "sturgeon",
+    "395": "gar, garfish, garpike, billfish, Lepisosteus osseus",
+    "396": "lionfish",
+    "397": "puffer, pufferfish, blowfish, globefish",
+    "398": "abacus",
+    "399": "abaya",
+    "400": "academic gown, academic robe, judge robe",
+    "401": "accordion, piano accordion, squeeze box",
+    "402": "acoustic guitar",
+    "403": "aircraft carrier, carrier, flattop, attack aircraft carrier",
+    "404": "airliner",
+    "405": "airship, dirigible",
+    "406": "altar",
+    "407": "ambulance",
+    "408": "amphibian, amphibious vehicle",
+    "409": "analog clock",
+    "410": "apiary, bee house",
+    "411": "apron",
+    "412": "ashcan, trash can, garbage can, wastebin, ash bin, ash-bin, ashbin, dustbin, trash barrel, trash bin",
+    "413": "assault rifle, assault gun",
+    "414": "backpack, back pack, knapsack, packsack, rucksack, haversack",
+    "415": "bakery, bakeshop, bakehouse",
+    "416": "balance beam, beam",
+    "417": "balloon",
+    "418": "ballpoint, ballpoint pen, ballpen, Biro",
+    "419": "Band Aid",
+    "420": "banjo",
+    "421": "bannister, banister, balustrade, balusters, handrail",
+    "422": "barbell",
+    "423": "barber chair",
+    "424": "barbershop",
+    "425": "barn",
+    "426": "barometer",
+    "427": "barrel, cask",
+    "428": "barrow, garden cart, lawn cart, wheelbarrow",
+    "429": "baseball",
+    "430": "basketball",
+    "431": "bassinet",
+    "432": "bassoon",
+    "433": "bathing cap, swimming cap",
+    "434": "bath towel",
+    "435": "bathtub, bathing tub, bath, tub",
+    "436": "beach wagon, station wagon, wagon, estate car, beach waggon, station waggon, waggon",
+    "437": "beacon, lighthouse, beacon light, pharos",
+    "438": "beaker",
+    "439": "bearskin, busby, shako",
+    "440": "beer bottle",
+    "441": "beer glass",
+    "442": "bell cote, bell cot",
+    "443": "bib",
+    "444": "bicycle-built-for-two, tandem bicycle, tandem",
+    "445": "bikini, two-piece",
+    "446": "binder, ring-binder",
+    "447": "binoculars, field glasses, opera glasses",
+    "448": "birdhouse",
+    "449": "boathouse",
+    "450": "bobsled, bobsleigh, bob",
+    "451": "bolo tie, bolo, bola tie, bola",
+    "452": "bonnet, poke bonnet",
+    "453": "bookcase",
+    "454": "bookshop, bookstore, bookstall",
+    "455": "bottlecap",
+    "456": "bow",
+    "457": "bow tie, bow-tie, bowtie",
+    "458": "brass, memorial tablet, plaque",
+    "459": "brassiere, bra, bandeau",
+    "460": "breakwater, groin, groyne, mole, bulwark, seawall, jetty",
+    "461": "breastplate, aegis, egis",
+    "462": "broom",
+    "463": "bucket, pail",
+    "464": "buckle",
+    "465": "bulletproof vest",
+    "466": "bullet train, bullet",
+    "467": "butcher shop, meat market",
+    "468": "cab, hack, taxi, taxicab",
+    "469": "caldron, cauldron",
+    "470": "candle, taper, wax light",
+    "471": "cannon",
+    "472": "canoe",
+    "473": "can opener, tin opener",
+    "474": "cardigan",
+    "475": "car mirror",
+    "476": "carousel, carrousel, merry-go-round, roundabout, whirligig",
+    "477": "carpenters kit, tool kit",
+    "478": "carton",
+    "479": "car wheel",
+    "480": "cash machine, cash dispenser, automated teller machine, automatic teller machine, automated teller, automatic teller, ATM",
+    "481": "cassette",
+    "482": "cassette player",
+    "483": "castle",
+    "484": "catamaran",
+    "485": "CD player",
+    "486": "cello, violoncello",
+    "487": "cellular telephone, cellular phone, cellphone, cell, mobile phone",
+    "488": "chain",
+    "489": "chainlink fence",
+    "490": "chain mail, ring mail, mail, chain armor, chain armour, ring armor, ring armour",
+    "491": "chain saw, chainsaw",
+    "492": "chest",
+    "493": "chiffonier, commode",
+    "494": "chime, bell, gong",
+    "495": "china cabinet, china closet",
+    "496": "Christmas stocking",
+    "497": "church, church building",
+    "498": "cinema, movie theater, movie theatre, movie house, picture palace",
+    "499": "cleaver, meat cleaver, chopper",
+    "500": "cliff dwelling",
+    "501": "cloak",
+    "502": "clog, geta, patten, sabot",
+    "503": "cocktail shaker",
+    "504": "coffee mug",
+    "505": "coffeepot",
+    "506": "coil, spiral, volute, whorl, helix",
+    "507": "combination lock",
+    "508": "computer keyboard, keypad",
+    "509": "confectionery, confectionary, candy store",
+    "510": "container ship, containership, container vessel",
+    "511": "convertible",
+    "512": "corkscrew, bottle screw",
+    "513": "cornet, horn, trumpet, trump",
+    "514": "cowboy boot",
+    "515": "cowboy hat, ten-gallon hat",
+    "516": "cradle",
+    "517": "crane",
+    "518": "crash helmet",
+    "519": "crate",
+    "520": "crib, cot",
+    "521": "Crock Pot",
+    "522": "croquet ball",
+    "523": "crutch",
+    "524": "cuirass",
+    "525": "dam, dike, dyke",
+    "526": "desk",
+    "527": "desktop computer",
+    "528": "dial telephone, dial phone",
+    "529": "diaper, nappy, napkin",
+    "530": "digital clock",
+    "531": "digital watch",
+    "532": "dining table, board",
+    "533": "dishrag, dishcloth",
+    "534": "dishwasher, dish washer, dishwashing machine",
+    "535": "disk brake, disc brake",
+    "536": "dock, dockage, docking facility",
+    "537": "dogsled, dog sled, dog sleigh",
+    "538": "dome",
+    "539": "doormat, welcome mat",
+    "540": "drilling platform, offshore rig",
+    "541": "drum, membranophone, tympan",
+    "542": "drumstick",
+    "543": "dumbbell",
+    "544": "Dutch oven",
+    "545": "electric fan, blower",
+    "546": "electric guitar",
+    "547": "electric locomotive",
+    "548": "entertainment center",
+    "549": "envelope",
+    "550": "espresso maker",
+    "551": "face powder",
+    "552": "feather boa, boa",
+    "553": "file, file cabinet, filing cabinet",
+    "554": "fireboat",
+    "555": "fire engine, fire truck",
+    "556": "fire screen, fireguard",
+    "557": "flagpole, flagstaff",
+    "558": "flute, transverse flute",
+    "559": "folding chair",
+    "560": "football helmet",
+    "561": "forklift",
+    "562": "fountain",
+    "563": "fountain pen",
+    "564": "four-poster",
+    "565": "freight car",
+    "566": "French horn, horn",
+    "567": "frying pan, frypan, skillet",
+    "568": "fur coat",
+    "569": "garbage truck, dustcart",
+    "570": "gasmask, respirator, gas helmet",
+    "571": "gas pump, gasoline pump, petrol pump, island dispenser",
+    "572": "goblet",
+    "573": "go-kart",
+    "574": "golf ball",
+    "575": "golfcart, golf cart",
+    "576": "gondola",
+    "577": "gong, tam-tam",
+    "578": "gown",
+    "579": "grand piano, grand",
+    "580": "greenhouse, nursery, glasshouse",
+    "581": "grille, radiator grille",
+    "582": "grocery store, grocery, food market, market",
+    "583": "guillotine",
+    "584": "hair slide",
+    "585": "hair spray",
+    "586": "half track",
+    "587": "hammer",
+    "588": "hamper",
+    "589": "hand blower, blow dryer, blow drier, hair dryer, hair drier",
+    "590": "hand-held computer, hand-held microcomputer",
+    "591": "handkerchief, hankie, hanky, hankey",
+    "592": "hard disc, hard disk, fixed disk",
+    "593": "harmonica, mouth organ, harp, mouth harp",
+    "594": "harp",
+    "595": "harvester, reaper",
+    "596": "hatchet",
+    "597": "holster",
+    "598": "home theater, home theatre",
+    "599": "honeycomb",
+    "600": "hook, claw",
+    "601": "hoopskirt, crinoline",
+    "602": "horizontal bar, high bar",
+    "603": "horse cart, horse-cart",
+    "604": "hourglass",
+    "605": "iPod",
+    "606": "iron, smoothing iron",
+    "607": "jack-o-lantern",
+    "608": "jean, blue jean, denim",
+    "609": "jeep, landrover",
+    "610": "jersey, T-shirt, tee shirt",
+    "611": "jigsaw puzzle",
+    "612": "jinrikisha, ricksha, rickshaw",
+    "613": "joystick",
+    "614": "kimono",
+    "615": "knee pad",
+    "616": "knot",
+    "617": "lab coat, laboratory coat",
+    "618": "ladle",
+    "619": "lampshade, lamp shade",
+    "620": "laptop, laptop computer",
+    "621": "lawn mower, mower",
+    "622": "lens cap, lens cover",
+    "623": "letter opener, paper knife, paperknife",
+    "624": "library",
+    "625": "lifeboat",
+    "626": "lighter, light, igniter, ignitor",
+    "627": "limousine, limo",
+    "628": "liner, ocean liner",
+    "629": "lipstick, lip rouge",
+    "630": "Loafer",
+    "631": "lotion",
+    "632": "loudspeaker, speaker, speaker unit, loudspeaker system, speaker system",
+    "633": "loupe, jewelers loupe",
+    "634": "lumbermill, sawmill",
+    "635": "magnetic compass",
+    "636": "mailbag, postbag",
+    "637": "mailbox, letter box",
+    "638": "maillot",
+    "639": "maillot, tank suit",
+    "640": "manhole cover",
+    "641": "maraca",
+    "642": "marimba, xylophone",
+    "643": "mask",
+    "644": "matchstick",
+    "645": "maypole",
+    "646": "maze, labyrinth",
+    "647": "measuring cup",
+    "648": "medicine chest, medicine cabinet",
+    "649": "megalith, megalithic structure",
+    "650": "microphone, mike",
+    "651": "microwave, microwave oven",
+    "652": "military uniform",
+    "653": "milk can",
+    "654": "minibus",
+    "655": "miniskirt, mini",
+    "656": "minivan",
+    "657": "missile",
+    "658": "mitten",
+    "659": "mixing bowl",
+    "660": "mobile home, manufactured home",
+    "661": "Model T",
+    "662": "modem",
+    "663": "monastery",
+    "664": "monitor",
+    "665": "moped",
+    "666": "mortar",
+    "667": "mortarboard",
+    "668": "mosque",
+    "669": "mosquito net",
+    "670": "motor scooter, scooter",
+    "671": "mountain bike, all-terrain bike, off-roader",
+    "672": "mountain tent",
+    "673": "mouse, computer mouse",
+    "674": "mousetrap",
+    "675": "moving van",
+    "676": "muzzle",
+    "677": "nail",
+    "678": "neck brace",
+    "679": "necklace",
+    "680": "nipple",
+    "681": "notebook, notebook computer",
+    "682": "obelisk",
+    "683": "oboe, hautboy, hautbois",
+    "684": "ocarina, sweet potato",
+    "685": "odometer, hodometer, mileometer, milometer",
+    "686": "oil filter",
+    "687": "organ, pipe organ",
+    "688": "oscilloscope, scope, cathode-ray oscilloscope, CRO",
+    "689": "overskirt",
+    "690": "oxcart",
+    "691": "oxygen mask",
+    "692": "packet",
+    "693": "paddle, boat paddle",
+    "694": "paddlewheel, paddle wheel",
+    "695": "padlock",
+    "696": "paintbrush",
+    "697": "pajama, pyjama, pjs, jammies",
+    "698": "palace",
+    "699": "panpipe, pandean pipe, syrinx",
+    "700": "paper towel",
+    "701": "parachute, chute",
+    "702": "parallel bars, bars",
+    "703": "park bench",
+    "704": "parking meter",
+    "705": "passenger car, coach, carriage",
+    "706": "patio, terrace",
+    "707": "pay-phone, pay-station",
+    "708": "pedestal, plinth, footstall",
+    "709": "pencil box, pencil case",
+    "710": "pencil sharpener",
+    "711": "perfume, essence",
+    "712": "Petri dish",
+    "713": "photocopier",
+    "714": "pick, plectrum, plectron",
+    "715": "pickelhaube",
+    "716": "picket fence, paling",
+    "717": "pickup, pickup truck",
+    "718": "pier",
+    "719": "piggy bank, penny bank",
+    "720": "pill bottle",
+    "721": "pillow",
+    "722": "ping-pong ball",
+    "723": "pinwheel",
+    "724": "pirate, pirate ship",
+    "725": "pitcher, ewer",
+    "726": "plane, carpenters plane, woodworking plane",
+    "727": "planetarium",
+    "728": "plastic bag",
+    "729": "plate rack",
+    "730": "plow, plough",
+    "731": "plunger, plumbers helper",
+    "732": "Polaroid camera, Polaroid Land camera",
+    "733": "pole",
+    "734": "police van, police wagon, paddy wagon, patrol wagon, wagon, black Maria",
+    "735": "poncho",
+    "736": "pool table, billiard table, snooker table",
+    "737": "pop bottle, soda bottle",
+    "738": "pot, flowerpot",
+    "739": "potters wheel",
+    "740": "power drill",
+    "741": "prayer rug, prayer mat",
+    "742": "printer",
+    "743": "prison, prison house",
+    "744": "projectile, missile",
+    "745": "projector",
+    "746": "puck, hockey puck",
+    "747": "punching bag, punch bag, punching ball, punchball",
+    "748": "purse",
+    "749": "quill, quill pen",
+    "750": "quilt, comforter, comfort, puff",
+    "751": "racer, race car, racing car",
+    "752": "racket, racquet",
+    "753": "radiator",
+    "754": "radio, wireless",
+    "755": "radio telescope, radio reflector",
+    "756": "rain barrel",
+    "757": "recreational vehicle, RV, R.V.",
+    "758": "reel",
+    "759": "reflex camera",
+    "760": "refrigerator, icebox",
+    "761": "remote control, remote",
+    "762": "restaurant, eating house, eating place, eatery",
+    "763": "revolver, six-gun, six-shooter",
+    "764": "rifle",
+    "765": "rocking chair, rocker",
+    "766": "rotisserie",
+    "767": "rubber eraser, rubber, pencil eraser",
+    "768": "rugby ball",
+    "769": "rule, ruler",
+    "770": "running shoe",
+    "771": "safe",
+    "772": "safety pin",
+    "773": "saltshaker, salt shaker",
+    "774": "sandal",
+    "775": "sarong",
+    "776": "sax, saxophone",
+    "777": "scabbard",
+    "778": "scale, weighing machine",
+    "779": "school bus",
+    "780": "schooner",
+    "781": "scoreboard",
+    "782": "screen, CRT screen",
+    "783": "screw",
+    "784": "screwdriver",
+    "785": "seat belt, seatbelt",
+    "786": "sewing machine",
+    "787": "shield, buckler",
+    "788": "shoe shop, shoe-shop, shoe store",
+    "789": "shoji",
+    "790": "shopping basket",
+    "791": "shopping cart",
+    "792": "shovel",
+    "793": "shower cap",
+    "794": "shower curtain",
+    "795": "ski",
+    "796": "ski mask",
+    "797": "sleeping bag",
+    "798": "slide rule, slipstick",
+    "799": "sliding door",
+    "800": "slot, one-armed bandit",
+    "801": "snorkel",
+    "802": "snowmobile",
+    "803": "snowplow, snowplough",
+    "804": "soap dispenser",
+    "805": "soccer ball",
+    "806": "sock",
+    "807": "solar dish, solar collector, solar furnace",
+    "808": "sombrero",
+    "809": "soup bowl",
+    "810": "space bar",
+    "811": "space heater",
+    "812": "space shuttle",
+    "813": "spatula",
+    "814": "speedboat",
+    "815": "spider web, spiders web",
+    "816": "spindle",
+    "817": "sports car, sport car",
+    "818": "spotlight, spot",
+    "819": "stage",
+    "820": "steam locomotive",
+    "821": "steel arch bridge",
+    "822": "steel drum",
+    "823": "stethoscope",
+    "824": "stole",
+    "825": "stone wall",
+    "826": "stopwatch, stop watch",
+    "827": "stove",
+    "828": "strainer",
+    "829": "streetcar, tram, tramcar, trolley, trolley car",
+    "830": "stretcher",
+    "831": "studio couch, day bed",
+    "832": "stupa, tope",
+    "833": "submarine, pigboat, sub, U-boat",
+    "834": "suit, suit of clothes",
+    "835": "sundial",
+    "836": "sunglass",
+    "837": "sunglasses, dark glasses, shades",
+    "838": "sunscreen, sunblock, sun blocker",
+    "839": "suspension bridge",
+    "840": "swab, swob, mop",
+    "841": "sweatshirt",
+    "842": "swimming trunks, bathing trunks",
+    "843": "swing",
+    "844": "switch, electric switch, electrical switch",
+    "845": "syringe",
+    "846": "table lamp",
+    "847": "tank, army tank, armored combat vehicle, armoured combat vehicle",
+    "848": "tape player",
+    "849": "teapot",
+    "850": "teddy, teddy bear",
+    "851": "television, television system",
+    "852": "tennis ball",
+    "853": "thatch, thatched roof",
+    "854": "theater curtain, theatre curtain",
+    "855": "thimble",
+    "856": "thresher, thrasher, threshing machine",
+    "857": "throne",
+    "858": "tile roof",
+    "859": "toaster",
+    "860": "tobacco shop, tobacconist shop, tobacconist",
+    "861": "toilet seat",
+    "862": "torch",
+    "863": "totem pole",
+    "864": "tow truck, tow car, wrecker",
+    "865": "toyshop",
+    "866": "tractor",
+    "867": "trailer truck, tractor trailer, trucking rig, rig, articulated lorry, semi",
+    "868": "tray",
+    "869": "trench coat",
+    "870": "tricycle, trike, velocipede",
+    "871": "trimaran",
+    "872": "tripod",
+    "873": "triumphal arch",
+    "874": "trolleybus, trolley coach, trackless trolley",
+    "875": "trombone",
+    "876": "tub, vat",
+    "877": "turnstile",
+    "878": "typewriter keyboard",
+    "879": "umbrella",
+    "880": "unicycle, monocycle",
+    "881": "upright, upright piano",
+    "882": "vacuum, vacuum cleaner",
+    "883": "vase",
+    "884": "vault",
+    "885": "velvet",
+    "886": "vending machine",
+    "887": "vestment",
+    "888": "viaduct",
+    "889": "violin, fiddle",
+    "890": "volleyball",
+    "891": "waffle iron",
+    "892": "wall clock",
+    "893": "wallet, billfold, notecase, pocketbook",
+    "894": "wardrobe, closet, press",
+    "895": "warplane, military plane",
+    "896": "washbasin, handbasin, washbowl, lavabo, wash-hand basin",
+    "897": "washer, automatic washer, washing machine",
+    "898": "water bottle",
+    "899": "water jug",
+    "900": "water tower",
+    "901": "whiskey jug",
+    "902": "whistle",
+    "903": "wig",
+    "904": "window screen",
+    "905": "window shade",
+    "906": "Windsor tie",
+    "907": "wine bottle",
+    "908": "wing",
+    "909": "wok",
+    "910": "wooden spoon",
+    "911": "wool, woolen, woollen",
+    "912": "worm fence, snake fence, snake-rail fence, Virginia fence",
+    "913": "wreck",
+    "914": "yawl",
+    "915": "yurt",
+    "916": "web site, website, internet site, site",
+    "917": "comic book",
+    "918": "crossword puzzle, crossword",
+    "919": "street sign",
+    "920": "traffic light, traffic signal, stoplight",
+    "921": "book jacket, dust cover, dust jacket, dust wrapper",
+    "922": "menu",
+    "923": "plate",
+    "924": "guacamole",
+    "925": "consomme",
+    "926": "hot pot, hotpot",
+    "927": "trifle",
+    "928": "ice cream, icecream",
+    "929": "ice lolly, lolly, lollipop, popsicle",
+    "930": "French loaf",
+    "931": "bagel, beigel",
+    "932": "pretzel",
+    "933": "cheeseburger",
+    "934": "hotdog, hot dog, red hot",
+    "935": "mashed potato",
+    "936": "head cabbage",
+    "937": "broccoli",
+    "938": "cauliflower",
+    "939": "zucchini, courgette",
+    "940": "spaghetti squash",
+    "941": "acorn squash",
+    "942": "butternut squash",
+    "943": "cucumber, cuke",
+    "944": "artichoke, globe artichoke",
+    "945": "bell pepper",
+    "946": "cardoon",
+    "947": "mushroom",
+    "948": "Granny Smith",
+    "949": "strawberry",
+    "950": "orange",
+    "951": "lemon",
+    "952": "fig",
+    "953": "pineapple, ananas",
+    "954": "banana",
+    "955": "jackfruit, jak, jack",
+    "956": "custard apple",
+    "957": "pomegranate",
+    "958": "hay",
+    "959": "carbonara",
+    "960": "chocolate sauce, chocolate syrup",
+    "961": "dough",
+    "962": "meat loaf, meatloaf",
+    "963": "pizza, pizza pie",
+    "964": "potpie",
+    "965": "burrito",
+    "966": "red wine",
+    "967": "espresso",
+    "968": "cup",
+    "969": "eggnog",
+    "970": "alp",
+    "971": "bubble",
+    "972": "cliff, drop, drop-off",
+    "973": "coral reef",
+    "974": "geyser",
+    "975": "lakeside, lakeshore",
+    "976": "promontory, headland, head, foreland",
+    "977": "sandbar, sand bar",
+    "978": "seashore, coast, seacoast, sea-coast",
+    "979": "valley, vale",
+    "980": "volcano",
+    "981": "ballplayer, baseball player",
+    "982": "groom, bridegroom",
+    "983": "scuba diver",
+    "984": "rapeseed",
+    "985": "daisy",
+    "986": "yellow ladys slipper, yellow lady-slipper, Cypripedium calceolus, Cypripedium parviflorum",
+    "987": "corn",
+    "988": "acorn",
+    "989": "hip, rose hip, rosehip",
+    "990": "buckeye, horse chestnut, conker",
+    "991": "coral fungus",
+    "992": "agaric",
+    "993": "gyromitra",
+    "994": "stinkhorn, carrion fungus",
+    "995": "earthstar",
+    "996": "hen-of-the-woods, hen of the woods, Polyporus frondosus, Grifola frondosa",
+    "997": "bolete",
+    "998": "ear, spike, capitulum",
+    "999": "toilet tissue, toilet paper, bathroom tissue"
+  }
+}

iMF-XL-2/pipeline.py

@@ -0,0 +1,275 @@
+# Copyright 2026 The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+"""Hub custom pipeline: IMFPipeline.
+
+Load with native Hugging Face diffusers and trust_remote_code=True.
+"""
+
+from __future__ import annotations
+
+import json
+from pathlib import Path
+from typing import Dict, List, Optional, Tuple, Union
+
+import torch
+from diffusers.image_processor import VaeImageProcessor
+from diffusers.pipelines.pipeline_utils import DiffusionPipeline, ImagePipelineOutput
+from diffusers.schedulers import FlowMatchEulerDiscreteScheduler
+from diffusers.utils.torch_utils import randn_tensor
+
+
+def _set_imf_timesteps(
+    scheduler: FlowMatchEulerDiscreteScheduler,
+    num_inference_steps: int,
+    device: torch.device,
+) -> torch.Tensor:
+    flow_sigmas = torch.linspace(1.0, 0.0, num_inference_steps + 1, device=device, dtype=torch.float32)
+    scheduler.set_timesteps(sigmas=flow_sigmas.tolist(), device=device)
+    return flow_sigmas
+
+
+class IMFPipeline(DiffusionPipeline):
+    r"""
+    Pipeline for ImageNet class-conditional generation with Improved Mean Flows (iMF).
+
+    Parameters:
+        transformer ([`IMFTransformer2DModel`]):
+            Class-conditioned iMF transformer that predicts mean-flow velocity.
+        scheduler ([`FlowMatchEulerDiscreteScheduler`]):
+            Flow-matching Euler scheduler.
+        vae ([`AutoencoderKL`]):
+            Variational autoencoder used to decode transformer latents to pixels.
+        id2label (`dict[int, str]`, *optional*):
+            ImageNet class id to English label mapping.
+    """
+
+    model_cpu_offload_seq = "transformer->vae"
+
+    def __init__(
+        self,
+        transformer,
+        scheduler,
+        vae,
+        id2label: Optional[Dict[Union[int, str], str]] = None,
+    ):
+        super().__init__()
+        if scheduler is None:
+            scheduler = FlowMatchEulerDiscreteScheduler(
+                num_train_timesteps=1000,
+                shift=1.0,
+                stochastic_sampling=False,
+            )
+        self.register_modules(transformer=transformer, scheduler=scheduler, vae=vae)
+        self.vae_scale_factor = 2 ** (len(self.vae.config.block_out_channels) - 1)
+        self.image_processor = VaeImageProcessor(vae_scale_factor=self.vae_scale_factor)
+        self._id2label = self._normalize_id2label(id2label)
+        self.labels = self._build_label2id(self._id2label)
+        self._labels_loaded_from_model_index = bool(self._id2label)
+
+    def _ensure_labels_loaded(self) -> None:
+        if self._labels_loaded_from_model_index:
+            return
+        loaded = self._read_id2label_from_model_index(getattr(self.config, "_name_or_path", None))
+        if loaded:
+            self._id2label = loaded
+            self.labels = self._build_label2id(self._id2label)
+        self._labels_loaded_from_model_index = True
+
+    @staticmethod
+    def _normalize_id2label(id2label: Optional[Dict[Union[int, str], str]]) -> Dict[int, str]:
+        if not id2label:
+            return {}
+        return {int(key): value for key, value in id2label.items()}
+
+    @staticmethod
+    def _read_id2label_from_model_index(variant_path: Optional[str]) -> Dict[int, str]:
+        if not variant_path:
+            return {}
+        model_index_path = Path(variant_path).resolve() / "model_index.json"
+        if not model_index_path.exists():
+            return {}
+        raw = json.loads(model_index_path.read_text(encoding="utf-8"))
+        id2label = raw.get("id2label")
+        if not isinstance(id2label, dict):
+            return {}
+        return {int(key): value for key, value in id2label.items()}
+
+    @staticmethod
+    def _build_label2id(id2label: Dict[int, str]) -> Dict[str, int]:
+        label2id: Dict[str, int] = {}
+        for class_id, value in id2label.items():
+            for synonym in value.split(","):
+                synonym = synonym.strip()
+                if synonym:
+                    label2id[synonym] = int(class_id)
+        return dict(sorted(label2id.items()))
+
+    @property
+    def id2label(self) -> Dict[int, str]:
+        self._ensure_labels_loaded()
+        return self._id2label
+
+    def get_label_ids(self, label: Union[str, List[str]]) -> List[int]:
+        self._ensure_labels_loaded()
+        if not self.labels:
+            raise ValueError("No labels loaded. Ensure `id2label` exists in model_index.json.")
+        labels = [label] if isinstance(label, str) else label
+        missing = [item for item in labels if item not in self.labels]
+        if missing:
+            preview = ", ".join(list(self.labels.keys())[:8])
+            raise ValueError(f"Unknown label(s): {missing}. Example valid labels: {preview}, ...")
+        return [self.labels[item] for item in labels]
+
+    def _normalize_class_labels(self, class_labels: Union[int, str, List[Union[int, str]]]) -> List[int]:
+        if isinstance(class_labels, int):
+            return [class_labels]
+        if isinstance(class_labels, str):
+            return self.get_label_ids(class_labels)
+        if class_labels and isinstance(class_labels[0], str):
+            return self.get_label_ids(class_labels)
+        return list(class_labels)
+
+    def _predict_velocity_u(
+        self,
+        latents: torch.Tensor,
+        timestep: torch.Tensor,
+        time_gap: torch.Tensor,
+        class_labels: torch.Tensor,
+        class_null: torch.Tensor,
+        guidance_scale: float,
+        guidance_interval_start: float,
+        guidance_interval_end: float,
+        do_classifier_free_guidance: bool,
+    ) -> torch.Tensor:
+        dtype = latents.dtype
+        timestep = timestep.to(device=latents.device, dtype=dtype)
+        time_gap = time_gap.to(device=latents.device, dtype=dtype)
+
+        if do_classifier_free_guidance:
+            latents_in = torch.cat([latents, latents], dim=0)
+            labels = torch.cat([class_labels, class_null], dim=0)
+            omega = torch.tensor([guidance_scale, 1.0], device=latents.device, dtype=dtype)
+            t_min = torch.tensor([guidance_interval_start, 0.0], device=latents.device, dtype=dtype)
+            t_max = torch.tensor([guidance_interval_end, 1.0], device=latents.device, dtype=dtype)
+            batch = latents.shape[0]
+            timestep_in = timestep.reshape(1).repeat(2 * batch)
+            time_gap_in = time_gap.reshape(1).repeat(2 * batch)
+            omega = omega.repeat(batch)
+            t_min = t_min.repeat(batch)
+            t_max = t_max.repeat(batch)
+        else:
+            latents_in = latents
+            labels = class_labels
+            batch = latents.shape[0]
+            timestep_in = timestep.reshape(1).repeat(batch)
+            time_gap_in = time_gap.reshape(1).repeat(batch)
+            omega = torch.full((batch,), guidance_scale, device=latents.device, dtype=dtype)
+            t_min = torch.full((batch,), guidance_interval_start, device=latents.device, dtype=dtype)
+            t_max = torch.full((batch,), guidance_interval_end, device=latents.device, dtype=dtype)
+
+        outputs = self.transformer(
+            sample=latents_in,
+            timestep=timestep_in,
+            class_labels=labels,
+            time_gap=time_gap_in,
+            guidance_scale=omega,
+            guidance_interval_start=t_min,
+            guidance_interval_end=t_max,
+            return_dict=True,
+        )
+        velocity_u = outputs.velocity_u
+
+        if not do_classifier_free_guidance:
+            return velocity_u
+
+        u_cond, u_uncond = velocity_u.chunk(2, dim=0)
+        return u_uncond + guidance_scale * (u_cond - u_uncond)
+
+    def decode_latents(self, latents: torch.Tensor, output_type: str = "pil"):
+        if output_type == "latent":
+            return latents
+
+        scaling_factor = self.vae.config.scaling_factor
+        latents = latents.to(device=self.vae.device, dtype=self.vae.dtype)
+        image = self.vae.decode(latents / scaling_factor).sample
+        if output_type == "pt":
+            return image
+        return self.image_processor.postprocess(image, output_type=output_type)
+
+    @torch.inference_mode()
+    def __call__(
+        self,
+        class_labels: Union[int, str, List[Union[int, str]]],
+        num_inference_steps: int = 1,
+        guidance_scale: float = 2.7,
+        guidance_interval_start: float = 0.1,
+        guidance_interval_end: float = 0.9,
+        generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
+        latents: Optional[torch.Tensor] = None,
+        output_type: str = "pil",
+        return_dict: bool = True,
+    ) -> Union[ImagePipelineOutput, Tuple]:
+        if output_type not in {"pil", "np", "pt", "latent"}:
+            raise ValueError("output_type must be one of: 'pil', 'np', 'pt', 'latent'.")
+
+        class_label_ids = self._normalize_class_labels(class_labels)
+        do_classifier_free_guidance = guidance_scale > 1.0
+        batch_size = len(class_label_ids)
+
+        image_size = int(self.transformer.config.sample_size)
+        channels = int(self.transformer.config.in_channels)
+        null_class_val = int(self.transformer.config.num_classes)
+
+        if latents is None:
+            latents = randn_tensor(
+                shape=(batch_size, channels, image_size, image_size),
+                generator=generator,
+                device=self._execution_device,
+                dtype=self.transformer.dtype,
+            )
+
+        class_labels_t = torch.tensor(class_label_ids, device=latents.device, dtype=torch.long).reshape(-1)
+        class_labels_t = class_labels_t.clamp(0, null_class_val - 1)
+        class_null = torch.full_like(class_labels_t, null_class_val)
+
+        flow_sigmas = _set_imf_timesteps(self.scheduler, num_inference_steps, latents.device)
+
+        for i in self.progress_bar(range(num_inference_steps)):
+            t = flow_sigmas[i]
+            t_next = flow_sigmas[i + 1]
+            time_gap = t - t_next
+            velocity_u = self._predict_velocity_u(
+                latents,
+                t,
+                time_gap,
+                class_labels_t,
+                class_null,
+                guidance_scale,
+                guidance_interval_start,
+                guidance_interval_end,
+                do_classifier_free_guidance,
+            )
+            latents = self.scheduler.step(velocity_u, self.scheduler.timesteps[i], latents).prev_sample
+
+        images = self.decode_latents(latents, output_type=output_type)
+
+        self.maybe_free_model_hooks()
+
+        if not return_dict:
+            return (images,)
+        return ImagePipelineOutput(images=images)
+
+
+IMFPipelineOutput = ImagePipelineOutput

iMF-XL-2/scheduler/scheduler_config.json

@@ -0,0 +1,7 @@
+{
+  "_class_name": "FlowMatchEulerDiscreteScheduler",
+  "_diffusers_version": "0.36.0",
+  "num_train_timesteps": 1000,
+  "shift": 1.0,
+  "stochastic_sampling": false
+}

iMF-XL-2/transformer/config.json

@@ -0,0 +1,23 @@
+{
+  "_class_name": "IMFTransformer2DModel",
+  "_diffusers_version": "0.38.0",
+  "aux_head_depth": 8,
+  "depth": 48,
+  "embedding_init_constant": 1.0,
+  "eval_mode": true,
+  "hidden_size": 1024,
+  "in_channels": 4,
+  "mlp_ratio": 2.6666666666666665,
+  "model_type": "iMF-XL/2",
+  "num_attention_heads": 16,
+  "num_cfg_tokens": 4,
+  "num_class_embeds": null,
+  "num_class_tokens": 8,
+  "num_classes": 1000,
+  "num_interval_tokens": 2,
+  "num_time_tokens": 4,
+  "patch_size": 2,
+  "sample_size": 32,
+  "token_init_constant": 1.0,
+  "weight_init_constant": 0.32
+}

iMF-XL-2/transformer/diffusion_pytorch_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:315b4c69c4fd7d9d6308d450499b8d95485e1d802c93413bb7ca1bb831c9cdc4
+size 2441803104

iMF-XL-2/transformer/transformer_imf.py

@@ -0,0 +1,586 @@
+from __future__ import annotations
+
+import argparse
+import math
+from collections.abc import Mapping
+from dataclasses import dataclass
+from functools import partial
+from math import sqrt
+from typing import Dict, Literal, Optional, Tuple, Union
+
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+from diffusers.configuration_utils import ConfigMixin, register_to_config
+from diffusers.models.modeling_utils import ModelMixin
+from diffusers.utils import BaseOutput
+
+
+IMF_PRESET_CONFIGS: Dict[str, Dict[str, object]] = {
+    "iMF-B/2": {
+        "sample_size": 32,
+        "patch_size": 2,
+        "hidden_size": 768,
+        "depth": 12,
+        "num_attention_heads": 12,
+        "aux_head_depth": 8,
+    },
+    "iMF-M/2": {
+        "sample_size": 32,
+        "patch_size": 2,
+        "hidden_size": 768,
+        "depth": 24,
+        "num_attention_heads": 12,
+        "aux_head_depth": 8,
+    },
+    "iMF-L/2": {
+        "sample_size": 32,
+        "patch_size": 2,
+        "hidden_size": 1024,
+        "depth": 32,
+        "num_attention_heads": 16,
+        "aux_head_depth": 8,
+    },
+    "iMF-XL/2": {
+        "sample_size": 32,
+        "patch_size": 2,
+        "hidden_size": 1024,
+        "depth": 48,
+        "num_attention_heads": 16,
+        "aux_head_depth": 8,
+    },
+}
+
+
+@dataclass
+class IMFTransformer2DModelOutput(BaseOutput):
+    velocity_u: torch.Tensor
+    velocity_v: Optional[torch.Tensor] = None
+
+
+def remap_legacy_state_dict(state_dict: Dict[str, torch.Tensor]) -> Dict[str, torch.Tensor]:
+    """Map legacy iMF / Flax-style keys to native IMFTransformer2DModel keys."""
+    remapped: Dict[str, torch.Tensor] = {}
+    for key, value in state_dict.items():
+        new_key = key
+        for prefix in ("net.", "transformer."):
+            if new_key.startswith(prefix):
+                new_key = new_key[len(prefix) :]
+                break
+        new_key = new_key.replace(".kernel", ".weight")
+        new_key = new_key.replace("._flax_linear.", ".linear.")
+        new_key = new_key.replace("._flax_embedding.", ".embedding.")
+        remapped[new_key] = value
+    return remapped
+
+
+def config_from_legacy(config: Dict[str, object]) -> Dict[str, object]:
+    model_type = config.get("model_type") or config.get("model_str") or config.get("model_name")
+    if model_type not in IMF_PRESET_CONFIGS:
+        raise ValueError(f"Unknown iMF preset '{model_type}'. Known: {list(IMF_PRESET_CONFIGS)}")
+
+    preset = dict(IMF_PRESET_CONFIGS[model_type])
+    preset["num_classes"] = int(config.get("num_class_embeds") or config.get("num_classes") or 1000)
+    preset["model_type"] = model_type
+    if config.get("sample_size") is not None:
+        preset["sample_size"] = int(config["sample_size"])
+    if config.get("in_channels") is not None:
+        preset["in_channels"] = int(config["in_channels"])
+    return preset
+
+
+def _unsqueeze(t: torch.Tensor, dim: int) -> torch.Tensor:
+    return t.unsqueeze(dim)
+
+
+class _ScaledLinear(nn.Module):
+    def __init__(
+        self,
+        in_features: int,
+        out_features: int,
+        bias: bool = True,
+        weight_init: str = "scaled_variance",
+        init_constant: float = 1.0,
+        bias_init: str = "zeros",
+    ):
+        super().__init__()
+        self.linear = nn.Linear(in_features, out_features, bias=bias)
+        if weight_init == "scaled_variance":
+            std = init_constant / sqrt(in_features)
+            nn.init.normal_(self.linear.weight, std=std)
+        elif weight_init == "zeros":
+            nn.init.zeros_(self.linear.weight)
+        else:
+            raise ValueError(f"Invalid weight_init: {weight_init}")
+        if bias:
+            if bias_init == "zeros":
+                nn.init.zeros_(self.linear.bias)
+            else:
+                raise ValueError(f"Invalid bias_init: {bias_init}")
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        return self.linear(x)
+
+
+class _ScaledEmbedding(nn.Module):
+    def __init__(
+        self,
+        num_embeddings: int,
+        embedding_dim: int,
+        weight_init: str = "scaled_variance",
+        init_constant: float = 1.0,
+    ):
+        super().__init__()
+        self.embedding = nn.Embedding(num_embeddings, embedding_dim)
+        if weight_init == "scaled_variance":
+            std = init_constant / sqrt(embedding_dim)
+            nn.init.normal_(self.embedding.weight, std=std)
+        else:
+            raise ValueError(f"Invalid weight_init: {weight_init}")
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        return self.embedding(x)
+
+
+class _RMSNorm(nn.Module):
+    def __init__(self, dim: int, eps: float = 1e-6):
+        super().__init__()
+        self.weight = nn.Parameter(torch.ones(dim))
+        self.eps = eps
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        mean_square = torch.mean(torch.square(x), dim=-1, keepdim=True)
+        output = x * torch.rsqrt(mean_square + self.eps)
+        return output.to(x.dtype) * self.weight
+
+
+class _SwiGLUMlp(nn.Module):
+    def __init__(
+        self,
+        in_features: int,
+        hidden_features: int,
+        weight_init: str = "scaled_variance",
+        weight_init_constant: float = 1.0,
+    ):
+        super().__init__()
+        init_kwargs = dict(bias=False, weight_init=weight_init, init_constant=weight_init_constant)
+        self.w1 = _ScaledLinear(in_features, hidden_features, **init_kwargs)
+        self.w3 = _ScaledLinear(in_features, hidden_features, **init_kwargs)
+        self.w2 = _ScaledLinear(hidden_features, in_features, **init_kwargs)
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        return self.w2(F.silu(self.w1(x)) * self.w3(x))
+
+
+class IMFTimestepEmbedder(nn.Module):
+    def __init__(
+        self,
+        hidden_size: int,
+        frequency_embedding_size: int = 256,
+        init_constant: float = 1.0,
+    ):
+        super().__init__()
+        self.mlp = nn.Sequential(
+            _ScaledLinear(frequency_embedding_size, hidden_size, bias=True, init_constant=init_constant),
+            nn.SiLU(),
+            _ScaledLinear(hidden_size, hidden_size, bias=True, init_constant=init_constant),
+        )
+        self.frequency_embedding_size = frequency_embedding_size
+
+    @staticmethod
+    def timestep_embedding(t: torch.Tensor, dim: int, max_period: int = 10000) -> torch.Tensor:
+        half = dim // 2
+        freqs = torch.exp(
+            -math.log(max_period) * torch.arange(start=0, end=half, dtype=torch.float32, device=t.device) / half
+        )
+        args = t[:, None].float() * freqs[None]
+        embedding = torch.cat([torch.cos(args), torch.sin(args)], dim=-1)
+        if dim % 2:
+            embedding = torch.cat([embedding, torch.zeros_like(embedding[:, :1])], dim=-1)
+        return embedding
+
+    def forward(self, t: torch.Tensor) -> torch.Tensor:
+        embedding = self.timestep_embedding(t, self.frequency_embedding_size)
+        return self.mlp(embedding.to(dtype=next(self.mlp.parameters()).dtype))
+
+
+class IMFLabelEmbedder(nn.Module):
+    def __init__(self, num_classes: int, hidden_size: int, init_constant: float = 1.0):
+        super().__init__()
+        self.embedding_table = _ScaledEmbedding(num_classes + 1, hidden_size, init_constant=init_constant)
+
+    def forward(self, labels: torch.Tensor) -> torch.Tensor:
+        return self.embedding_table(labels)
+
+
+class IMFPatchEmbedder(nn.Module):
+    def __init__(self, input_size: int, patch_size: int, in_channels: int, hidden_size: int, bias: bool = True):
+        super().__init__()
+        self.patch_size = (patch_size, patch_size)
+        self.num_patches = (input_size // patch_size) ** 2
+        self.proj = nn.Conv2d(
+            in_channels,
+            hidden_size,
+            kernel_size=patch_size,
+            stride=patch_size,
+            bias=bias,
+        )
+        kh = kw = patch_size
+        fan_in = kh * kw * in_channels
+        fan_out = hidden_size
+        limit = math.sqrt(6.0 / (fan_in + fan_out))
+        nn.init.uniform_(self.proj.weight, -limit, limit)
+        if bias:
+            nn.init.zeros_(self.proj.bias)
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        x = self.proj(x)
+        return x.flatten(2).transpose(1, 2)
+
+
+def precompute_rope_freqs(dim: int, seq_len: int, theta: float = 10000.0, device: torch.device | None = None):
+    freqs = 1.0 / (theta ** (torch.arange(0, dim, 2, dtype=torch.float32, device=device) / dim))
+    positions = torch.arange(seq_len, dtype=torch.float32, device=device)
+    freqs_cis = torch.outer(positions, freqs)
+    return torch.complex(torch.cos(freqs_cis), torch.sin(freqs_cis))
+
+
+def apply_rotary_pos_emb(x: torch.Tensor, freqs_cis: torch.Tensor) -> torch.Tensor:
+    x_ = x.float().reshape(*x.shape[:-1], -1, 2)
+    x_complex = torch.view_as_complex(x_)
+    freqs_cis = freqs_cis.unsqueeze(0).unsqueeze(2).to(x.device)
+    x_rotated = torch.view_as_real(x_complex * freqs_cis).flatten(-2)
+    return x_rotated.to(dtype=x.dtype)
+
+
+class IMFRoPEAttention(nn.Module):
+    def __init__(self, hidden_size: int, num_heads: int, weight_init_constant: float = 0.32):
+        super().__init__()
+        self.hidden_size = hidden_size
+        self.num_heads = num_heads
+        self.head_dim = hidden_size // num_heads
+        init_kwargs = dict(bias=False, init_constant=weight_init_constant)
+        self.q_proj = _ScaledLinear(hidden_size, hidden_size, **init_kwargs)
+        self.k_proj = _ScaledLinear(hidden_size, hidden_size, **init_kwargs)
+        self.v_proj = _ScaledLinear(hidden_size, hidden_size, **init_kwargs)
+        self.out_proj = _ScaledLinear(hidden_size, hidden_size, **init_kwargs)
+        self.q_norm = _RMSNorm(self.head_dim)
+        self.k_norm = _RMSNorm(self.head_dim)
+
+    def forward(self, x: torch.Tensor, rope_freqs: torch.Tensor) -> torch.Tensor:
+        batch, seq_len, _ = x.shape
+        q = self.q_proj(x).reshape(batch, seq_len, self.num_heads, self.head_dim)
+        k = self.k_proj(x).reshape(batch, seq_len, self.num_heads, self.head_dim)
+        v = self.v_proj(x).reshape(batch, seq_len, self.num_heads, self.head_dim)
+        q = apply_rotary_pos_emb(self.q_norm(q), rope_freqs)
+        k = apply_rotary_pos_emb(self.k_norm(k), rope_freqs)
+        query = q / math.sqrt(self.head_dim)
+        attn_weights = torch.einsum("bqhd,bkhd->bhqk", query, k)
+        attn_weights = F.softmax(attn_weights, dim=-1, dtype=torch.float32).to(v.dtype)
+        attn = torch.einsum("bhqk,bkhd->bqhd", attn_weights, v)
+        return self.out_proj(attn.reshape(batch, seq_len, self.hidden_size))
+
+
+class IMFTransformerBlock(nn.Module):
+    def __init__(self, hidden_size: int, num_heads: int, mlp_ratio: float, weight_init_constant: float):
+        super().__init__()
+        self.norm1 = _RMSNorm(hidden_size)
+        self.attn = IMFRoPEAttention(hidden_size, num_heads, weight_init_constant=weight_init_constant)
+        self.norm2 = _RMSNorm(hidden_size)
+        mlp_hidden_dim = int(hidden_size * mlp_ratio)
+        self.mlp = _SwiGLUMlp(hidden_size, mlp_hidden_dim, weight_init_constant=weight_init_constant)
+        self.attn_scale = nn.Parameter(torch.zeros(hidden_size))
+        self.mlp_scale = nn.Parameter(torch.zeros(hidden_size))
+
+    def forward(self, x: torch.Tensor, rope_freqs: torch.Tensor) -> torch.Tensor:
+        x = x + self.attn(self.norm1(x), rope_freqs) * self.attn_scale
+        x = x + self.mlp(self.norm2(x)) * self.mlp_scale
+        return x
+
+
+class IMFFinalLayer(nn.Module):
+    def __init__(self, hidden_size: int, patch_size: int, out_channels: int):
+        super().__init__()
+        self.norm = _RMSNorm(hidden_size)
+        self.linear = _ScaledLinear(
+            hidden_size,
+            patch_size * patch_size * out_channels,
+            bias=True,
+            weight_init="zeros",
+            init_constant=1.0,
+        )
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        return self.linear(self.norm(x))
+
+
+class IMFTransformer2DModel(ModelMixin, ConfigMixin):
+    _supports_gradient_checkpointing = True
+
+    @register_to_config
+    def __init__(
+        self,
+        sample_size: int = 32,
+        patch_size: int = 2,
+        in_channels: int = 4,
+        hidden_size: int = 768,
+        depth: int = 12,
+        num_attention_heads: int = 12,
+        mlp_ratio: float = 8 / 3,
+        num_classes: int = 1000,
+        aux_head_depth: int = 8,
+        num_class_tokens: int = 8,
+        num_time_tokens: int = 4,
+        num_cfg_tokens: int = 4,
+        num_interval_tokens: int = 2,
+        token_init_constant: float = 1.0,
+        embedding_init_constant: float = 1.0,
+        weight_init_constant: float = 0.32,
+        eval_mode: bool = True,
+        model_type: str | None = None,
+        num_class_embeds: int | None = None,
+    ):
+        super().__init__()
+        if num_class_embeds is not None:
+            num_classes = int(num_class_embeds)
+        if model_type in IMF_PRESET_CONFIGS:
+            preset = IMF_PRESET_CONFIGS[model_type]
+            sample_size = int(preset["sample_size"])
+            patch_size = int(preset["patch_size"])
+            hidden_size = int(preset["hidden_size"])
+            depth = int(preset["depth"])
+            num_attention_heads = int(preset["num_attention_heads"])
+            aux_head_depth = int(preset["aux_head_depth"])
+
+        self.sample_size = sample_size
+        self.patch_size = patch_size
+        self.in_channels = in_channels
+        self.out_channels = in_channels
+        self.hidden_size = hidden_size
+        self.depth = depth
+        self.num_attention_heads = num_attention_heads
+        self.num_classes = num_classes
+        self.aux_head_depth = aux_head_depth
+        self.num_class_tokens = num_class_tokens
+        self.num_time_tokens = num_time_tokens
+        self.num_cfg_tokens = num_cfg_tokens
+        self.num_interval_tokens = num_interval_tokens
+        self.eval_mode = eval_mode
+        self.gradient_checkpointing = False
+
+        self.x_embedder = IMFPatchEmbedder(sample_size, patch_size, in_channels, hidden_size, bias=True)
+        embed_kwargs = dict(hidden_size=hidden_size, init_constant=embedding_init_constant)
+        self.h_embedder = IMFTimestepEmbedder(**embed_kwargs)
+        self.omega_embedder = IMFTimestepEmbedder(**embed_kwargs)
+        self.cfg_t_start_embedder = IMFTimestepEmbedder(**embed_kwargs)
+        self.cfg_t_end_embedder = IMFTimestepEmbedder(**embed_kwargs)
+        self.y_embedder = IMFLabelEmbedder(num_classes, hidden_size, init_constant=embedding_init_constant)
+
+        token_std = token_init_constant / math.sqrt(hidden_size)
+        self.time_tokens = nn.Parameter(torch.randn(num_time_tokens, hidden_size) * token_std)
+        self.class_tokens = nn.Parameter(torch.randn(num_class_tokens, hidden_size) * token_std)
+        self.omega_tokens = nn.Parameter(torch.randn(num_cfg_tokens, hidden_size) * token_std)
+        self.t_min_tokens = nn.Parameter(torch.randn(num_interval_tokens, hidden_size) * token_std)
+        self.t_max_tokens = nn.Parameter(torch.randn(num_interval_tokens, hidden_size) * token_std)
+
+        total_tokens = (
+            self.x_embedder.num_patches
+            + num_class_tokens
+            + num_cfg_tokens
+            + 2 * num_interval_tokens
+            + num_time_tokens
+        )
+        self.prefix_tokens = num_class_tokens + num_cfg_tokens + 2 * num_interval_tokens + num_time_tokens
+        self.head_dim = hidden_size // num_attention_heads
+        self.register_buffer(
+            "rope_freqs",
+            precompute_rope_freqs(self.head_dim, total_tokens),
+            persistent=False,
+        )
+
+        shared_depth = depth - aux_head_depth
+        block_kwargs = dict(
+            hidden_size=hidden_size,
+            num_heads=num_attention_heads,
+            mlp_ratio=mlp_ratio,
+            weight_init_constant=weight_init_constant,
+        )
+        self.shared_blocks = nn.ModuleList([IMFTransformerBlock(**block_kwargs) for _ in range(shared_depth)])
+        self.u_heads = nn.ModuleList([IMFTransformerBlock(**block_kwargs) for _ in range(aux_head_depth)])
+        self.v_heads = nn.ModuleList(
+            [IMFTransformerBlock(**block_kwargs) for _ in range(aux_head_depth if not eval_mode else 0)]
+        )
+        self.u_final_layer = IMFFinalLayer(hidden_size, patch_size, in_channels)
+        self.v_final_layer = IMFFinalLayer(hidden_size, patch_size, in_channels)
+
+    def unpatchify(self, x: torch.Tensor) -> torch.Tensor:
+        c = self.out_channels
+        p = self.patch_size
+        h = w = int(x.shape[1] ** 0.5)
+        x = x.reshape(x.shape[0], h, w, p, p, c)
+        x = torch.einsum("nhwpqc->nchpwq", x)
+        return x.reshape(x.shape[0], c, h * p, w * p)
+
+    def _build_sequence(
+        self,
+        sample: torch.Tensor,
+        time_gap: torch.Tensor,
+        guidance_scale: torch.Tensor,
+        guidance_interval_start: torch.Tensor,
+        guidance_interval_end: torch.Tensor,
+        class_labels: torch.Tensor,
+    ) -> torch.Tensor:
+        x_embed = self.x_embedder(sample)
+        h_embed = self.h_embedder(time_gap)
+        omega_embed = self.omega_embedder(1 - 1 / guidance_scale)
+        t_min_embed = self.cfg_t_start_embedder(guidance_interval_start)
+        t_max_embed = self.cfg_t_end_embedder(guidance_interval_end)
+        y_embed = self.y_embedder(class_labels)
+
+        time_tokens = self.time_tokens + _unsqueeze(h_embed, 1)
+        omega_tokens = self.omega_tokens + _unsqueeze(omega_embed, 1)
+        t_min_tokens = self.t_min_tokens + _unsqueeze(t_min_embed, 1)
+        t_max_tokens = self.t_max_tokens + _unsqueeze(t_max_embed, 1)
+        class_tokens = self.class_tokens + _unsqueeze(y_embed, 1)
+
+        return torch.cat(
+            [class_tokens, omega_tokens, t_min_tokens, t_max_tokens, time_tokens, x_embed],
+            dim=1,
+        )
+
+    def forward(
+        self,
+        sample: torch.Tensor,
+        timestep: torch.Tensor,
+        class_labels: torch.Tensor,
+        time_gap: Optional[torch.Tensor] = None,
+        guidance_scale: Optional[torch.Tensor] = None,
+        guidance_interval_start: Optional[torch.Tensor] = None,
+        guidance_interval_end: Optional[torch.Tensor] = None,
+        return_dict: bool = True,
+    ) -> Union[IMFTransformer2DModelOutput, Tuple[torch.Tensor, ...]]:
+        batch_size = sample.shape[0]
+        timestep = self._expand_batch(timestep, batch_size, sample.device, sample.dtype)
+        if time_gap is None:
+            time_gap = timestep
+        else:
+            time_gap = self._expand_batch(time_gap, batch_size, sample.device, sample.dtype)
+        if guidance_scale is None:
+            guidance_scale = torch.ones(batch_size, device=sample.device, dtype=sample.dtype)
+        else:
+            guidance_scale = self._expand_batch(guidance_scale, batch_size, sample.device, sample.dtype)
+        if guidance_interval_start is None:
+            guidance_interval_start = torch.zeros(batch_size, device=sample.device, dtype=sample.dtype)
+        else:
+            guidance_interval_start = self._expand_batch(
+                guidance_interval_start, batch_size, sample.device, sample.dtype
+            )
+        if guidance_interval_end is None:
+            guidance_interval_end = torch.ones(batch_size, device=sample.device, dtype=sample.dtype)
+        else:
+            guidance_interval_end = self._expand_batch(
+                guidance_interval_end, batch_size, sample.device, sample.dtype
+            )
+
+        class_labels = class_labels.reshape(-1).long()
+        seq = self._build_sequence(
+            sample,
+            time_gap,
+            guidance_scale,
+            guidance_interval_start,
+            guidance_interval_end,
+            class_labels,
+        )
+
+        for block in self.shared_blocks:
+            if self.training and self.gradient_checkpointing:
+                seq = torch.utils.checkpoint.checkpoint(block, seq, self.rope_freqs, use_reentrant=False)
+            else:
+                seq = block(seq, self.rope_freqs)
+
+        u_seq = v_seq = seq
+        for block in self.u_heads:
+            u_seq = block(u_seq, self.rope_freqs)
+        for block in self.v_heads:
+            v_seq = block(v_seq, self.rope_freqs)
+
+        u_tokens = u_seq[:, self.prefix_tokens :]
+        velocity_u = self.unpatchify(self.u_final_layer(u_tokens))
+
+        velocity_v = None
+        if len(self.v_heads) > 0:
+            v_tokens = v_seq[:, self.prefix_tokens :]
+            velocity_v = self.unpatchify(self.v_final_layer(v_tokens))
+
+        if not return_dict:
+            if velocity_v is None:
+                return (velocity_u,)
+            return (velocity_u, velocity_v)
+        return IMFTransformer2DModelOutput(velocity_u=velocity_u, velocity_v=velocity_v)
+
+    @staticmethod
+    def _expand_batch(
+        value: torch.Tensor,
+        batch_size: int,
+        device: torch.device,
+        dtype: torch.dtype,
+    ) -> torch.Tensor:
+        value = torch.as_tensor(value, device=device, dtype=dtype)
+        if value.ndim == 0:
+            value = value.reshape(1).repeat(batch_size)
+        else:
+            value = value.reshape(-1)
+            if value.shape[0] == 1 and batch_size > 1:
+                value = value.repeat(batch_size)
+        return value
+
+    @classmethod
+    def from_imf_checkpoint(
+        cls,
+        checkpoint_path: str,
+        model_type: str | None = None,
+        map_location: str = "cpu",
+        strict: bool = True,
+        eval_mode: bool = True,
+    ) -> Tuple["IMFTransformer2DModel", Dict[str, object]]:
+        checkpoint = torch.load(checkpoint_path, map_location=map_location, weights_only=False)
+        if isinstance(checkpoint, Mapping) and "state_dict" in checkpoint:
+            state_dict = checkpoint["state_dict"]
+        elif isinstance(checkpoint, Mapping) and any(k.startswith("net.") for k in checkpoint):
+            state_dict = checkpoint
+        else:
+            state_dict = checkpoint
+
+        inferred_type = model_type
+        if inferred_type is None and isinstance(checkpoint, Mapping):
+            args = checkpoint.get("args")
+            if isinstance(args, argparse.Namespace):
+                inferred_type = getattr(args, "model", None) or getattr(args, "model_str", None)
+            elif isinstance(args, Mapping):
+                inferred_type = args.get("model") or args.get("model_str")
+
+        if inferred_type is None:
+            raise ValueError("model_type must be provided when loading a raw state dict checkpoint.")
+
+        config = dict(IMF_PRESET_CONFIGS[inferred_type])
+        config["model_type"] = inferred_type
+        config["eval_mode"] = eval_mode
+        model = cls(**config)
+        model.load_state_dict(remap_legacy_state_dict(state_dict), strict=strict)
+        metadata = {"checkpoint_path": checkpoint_path, "model_type": inferred_type}
+        return model, metadata
+
+    def to_imf_checkpoint(self, prefix: str = "net.") -> Dict[str, torch.Tensor]:
+        state: Dict[str, torch.Tensor] = {}
+        for key, value in self.state_dict().items():
+            if key == "rope_freqs":
+                continue
+            state[f"{prefix}{key}"] = value.detach().cpu()
+        return state
+
+    @property
+    def net(self):
+        return self
+
+
+IMFDiffusersModel = IMFTransformer2DModel

iMF-XL-2/vae/config.json

@@ -0,0 +1,38 @@
+{
+  "_class_name": "AutoencoderKL",
+  "_diffusers_version": "0.36.0",
+  "_name_or_path": "stabilityai/sd-vae-ft-mse",
+  "act_fn": "silu",
+  "block_out_channels": [
+    128,
+    256,
+    512,
+    512
+  ],
+  "down_block_types": [
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D"
+  ],
+  "force_upcast": true,
+  "in_channels": 3,
+  "latent_channels": 4,
+  "latents_mean": null,
+  "latents_std": null,
+  "layers_per_block": 2,
+  "mid_block_add_attention": true,
+  "norm_num_groups": 32,
+  "out_channels": 3,
+  "sample_size": 256,
+  "scaling_factor": 0.18215,
+  "shift_factor": null,
+  "up_block_types": [
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D"
+  ],
+  "use_post_quant_conv": true,
+  "use_quant_conv": true
+}

iMF-XL-2/vae/diffusion_pytorch_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a1d993488569e928462932c8c38a0760b874d166399b14414135bd9c42df5815
+size 334643276