Instructions to use IQuestLab/IQuest-Coder-V1-40B-Loop-Instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use IQuestLab/IQuest-Coder-V1-40B-Loop-Instruct with Transformers:

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

pipe = pipeline("text-generation", model="IQuestLab/IQuest-Coder-V1-40B-Loop-Instruct", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("IQuestLab/IQuest-Coder-V1-40B-Loop-Instruct", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use IQuestLab/IQuest-Coder-V1-40B-Loop-Instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "IQuestLab/IQuest-Coder-V1-40B-Loop-Instruct"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "IQuestLab/IQuest-Coder-V1-40B-Loop-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/IQuestLab/IQuest-Coder-V1-40B-Loop-Instruct

SGLang

How to use IQuestLab/IQuest-Coder-V1-40B-Loop-Instruct 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 "IQuestLab/IQuest-Coder-V1-40B-Loop-Instruct" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "IQuestLab/IQuest-Coder-V1-40B-Loop-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "IQuestLab/IQuest-Coder-V1-40B-Loop-Instruct" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "IQuestLab/IQuest-Coder-V1-40B-Loop-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use IQuestLab/IQuest-Coder-V1-40B-Loop-Instruct with Docker Model Runner:
```
docker model run hf.co/IQuestLab/IQuest-Coder-V1-40B-Loop-Instruct
```

Question about K2/V2 cache computation in prefill vs generation

#10

by kernelpool - opened Jan 2

Discussion

kernelpool

Jan 2

I'm trying to understand the caching behavior in modeling_iquestloopcoder.py and noticed a difference between prefill and generation:

Prefill (_forward_loop, lines 1072-1074):

hidden_states, gate_mean = decoder_layer.forward_loop2_mixed(...)
if use_cache and loop_idx == 2:
    hidden_states_normed = decoder_layer.input_layernorm(hidden_states)
    _, k2, v2 = decoder_layer.self_attn.get_qkv(hidden_states_normed, position_ids)

Here hidden_states is the layer OUTPUT (after attention + MLP).

Generation (forward_decode_loop2, line 630):

q2, k2, v2 = self.get_qkv(hidden_states, position_ids)

Here hidden_states is the layer INPUT (before attention).

Is the difference in prefill intentional, or should K2/V2 be computed from the same source in both paths? I ran some tests comparing both approaches against full recomputation and found that INPUT-based K2 matches exactly, while OUTPUT-based differs slightly. However, the practical impact seems minimal since the gates strongly favor global attention (~87%). I'm curious whether the difference is intentional or an oversight.

yxing-bj

IQuest org Jan 7

•

edited Jan 7

Sorry for the delayed reply. We had improved code clarity by refactoring.

kernelpool

Jan 7

Great, thanks for the clarification!

kernelpool changed discussion status to closed Jan 7

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment