Instructions to use YOYO-AI/QwQ-coder-32B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use YOYO-AI/QwQ-coder-32B with Transformers:

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

pipe = pipeline("text-generation", model="YOYO-AI/QwQ-coder-32B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("YOYO-AI/QwQ-coder-32B")
model = AutoModelForCausalLM.from_pretrained("YOYO-AI/QwQ-coder-32B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use YOYO-AI/QwQ-coder-32B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "YOYO-AI/QwQ-coder-32B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "YOYO-AI/QwQ-coder-32B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/YOYO-AI/QwQ-coder-32B

SGLang

How to use YOYO-AI/QwQ-coder-32B 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 "YOYO-AI/QwQ-coder-32B" \
    --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": "YOYO-AI/QwQ-coder-32B",
		"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 "YOYO-AI/QwQ-coder-32B" \
        --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": "YOYO-AI/QwQ-coder-32B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use YOYO-AI/QwQ-coder-32B with Docker Model Runner:
```
docker model run hf.co/YOYO-AI/QwQ-coder-32B
```

Short context? why?

by d00mus - opened Mar 16, 2025

Discussion

d00mus

Mar 16, 2025

Both the parent models have longer context, up to 128k but this one is only 32k. which is really disappointing..Is it possible to fix that?

YOYO-AI

Owner Mar 16, 2025

This comment has been hidden

d00mus

Mar 16, 2025

•

edited Mar 16, 2025

Just re-checked..Qwen says that both, QwQ and Coder-instruct have 128k context.. But you mentioned Just Coder(no instruct?).. Actually, "no instruct" also has 128k..So it's really should not be a problem..

YOYO-AI

Owner Mar 17, 2025

@d00mus The problem has been solved. Thank you for your feedback. The context has been changed to 128K!

@mradermacher Now that the problem of the context length has been solved, could you please provide the quantized version of the model again? Thank you so much for your help!

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