Instructions to use Qwen/Qwen3-32B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Qwen/Qwen3-32B with Transformers:

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

pipe = pipeline("text-generation", model="Qwen/Qwen3-32B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen3-32B")
model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-32B")
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]:]))

Inference
HuggingChat
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Qwen/Qwen3-32B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Qwen/Qwen3-32B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Qwen/Qwen3-32B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Qwen/Qwen3-32B

SGLang

How to use Qwen/Qwen3-32B 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 "Qwen/Qwen3-32B" \
    --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": "Qwen/Qwen3-32B",
		"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 "Qwen/Qwen3-32B" \
        --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": "Qwen/Qwen3-32B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Qwen/Qwen3-32B with Docker Model Runner:
```
docker model run hf.co/Qwen/Qwen3-32B
```

"/no_think" control is unstable

#16

by Smorty100 - opened Apr 29, 2025

Discussion

Smorty100

Apr 29, 2025

•

edited Apr 29, 2025

When passing "/no_think", "/no_thinking" or similar parameters into the system prompt, qwen3 generates no thoughts, as expected.

However!

The next conversation turn DOES include a thinking section.

so a conversation might look like this:

[
    {"role": "system", "content": "/no_think"},
    {"role": "user", "content": "hiii my sweetie qwen!"},
    {"role": "assistant", "content": "<think>\n\n</think>\nhi! how may i assist u today?"},
    {"role": "user", "content": "well, how do you do?"},
    {"role": "assistant", "content": "<think>\nOkay, the user said \"blablabla\". Let's break this down step by step [...]\n</think>\nim doin fine, thanks for askin"}
]

in the example, the first reponse was correctly without thought content, but the latter did have thinking content.

Here an instance of this happening in ollama on qwen3:4b (Q4K)

MrDevolver

Apr 29, 2025

I was dealing with stubborn thinking enabled despite specifically setting it to disabled in a roleplay scenario where I just wanted to see the direct response and nothing else. What I ended up doing was that I was prepending that <think>\n\n</think>\n into AI's response and it started giving me the direct answers. It's not ideal, but it works.

jesulo

Sep 25, 2025

Solved?

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