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

Libraries

How to use thirdeyeai/QwQ-32B-uncensored with Transformers:

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

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

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("thirdeyeai/QwQ-32B-uncensored")
model = AutoModelForCausalLM.from_pretrained("thirdeyeai/QwQ-32B-uncensored")
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 thirdeyeai/QwQ-32B-uncensored with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/thirdeyeai/QwQ-32B-uncensored

SGLang

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

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

report

by DataSoul - opened Mar 12, 2025

Discussion

DataSoul

Mar 12, 2025

It seems that "uncensored" didn't work.

sloshywings

thirdeye Ai org Apr 4, 2025

•

edited Apr 4, 2025

thanks, have to work on an updated method for uncensoring for this model. it still has removed the political bias from rlhf training. For example, if you had asked about tiananmen square the current qwq model does not provide an answer or flags as innapropriate.

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