Instructions to use Gryphe/MythoLogic-Mini-7b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Gryphe/MythoLogic-Mini-7b with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="Gryphe/MythoLogic-Mini-7b")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Gryphe/MythoLogic-Mini-7b")
model = AutoModelForCausalLM.from_pretrained("Gryphe/MythoLogic-Mini-7b")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Gryphe/MythoLogic-Mini-7b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Gryphe/MythoLogic-Mini-7b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Gryphe/MythoLogic-Mini-7b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Gryphe/MythoLogic-Mini-7b

SGLang

How to use Gryphe/MythoLogic-Mini-7b with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "Gryphe/MythoLogic-Mini-7b" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Gryphe/MythoLogic-Mini-7b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "Gryphe/MythoLogic-Mini-7b" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Gryphe/MythoLogic-Mini-7b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Gryphe/MythoLogic-Mini-7b with Docker Model Runner:
```
docker model run hf.co/Gryphe/MythoLogic-Mini-7b
```

MythoLogic-Mini-7b / README.md

Gryphe

Update README.md

43f822e almost 3 years ago

preview code

raw

history blame

1.61 kB

	---
	license: other
	language:
	- en
	---
	## Model details

	MythoLogic-Mini-7b can be considered the little brother in my Mytho series of models: [MythoLogic-13b](https://huggingface.co/Gryphe/MythoLogic-13b) and [MythoBoros-13b](https://huggingface.co/Gryphe/MythoBoros-13b)).

	Its Llama-2 core is powered by [Nous Hermes-2](https://huggingface.co/NousResearch/Nous-Hermes-llama-2-7b), which is further augmented by [Stable Beluga](https://huggingface.co/stabilityai/StableBeluga-7B) and a carefully distilled [Kimiko LoRa](https://huggingface.co/nRuaif/Kimiko_7B).

	Since 7B models tend to be less capable all-rounders, more emphasis was put on improving the roleplaying aspects for this gradient merge, of which various gradients were benchmarked before settling on the configuration shown below.

	![](MythoLogic-Mini-7b.png)

	In technical terms, the Hermes-2 core starts at 90% strength before fading away completely at the 12th layer level, where Stable Beluga (and Kimiko) handle the more intricate linguistic aspects.

	Quantized models are available from TheBloke: [GGML](https://huggingface.co/TheBloke/MythoLogic-Mini-7b-GGML) - [GPTQ](https://huggingface.co/TheBloke/MythoLogic-Mini-7b-GPTQ) (You're the best!)

	## Prompt Format

	Due to its Hermes-2 core this model works best with Alpaca formatting, so for optimal model performance, use:
	```
	<System prompt/Character Card>

	### Instruction:
	Your instruction or question here.
	For roleplay purposes, I suggest the following - Write <CHAR NAME>'s next reply in a chat between <YOUR NAME> and <CHAR NAME>. Write a single reply only.

	### Response:
	```