Instructions to use nanzhang/QuantLRM-R1-Qwen-32B-3-bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use nanzhang/QuantLRM-R1-Qwen-32B-3-bit with Transformers:

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

pipe = pipeline("text-generation", model="nanzhang/QuantLRM-R1-Qwen-32B-3-bit")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("nanzhang/QuantLRM-R1-Qwen-32B-3-bit")
model = AutoModelForCausalLM.from_pretrained("nanzhang/QuantLRM-R1-Qwen-32B-3-bit")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use nanzhang/QuantLRM-R1-Qwen-32B-3-bit with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "nanzhang/QuantLRM-R1-Qwen-32B-3-bit"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "nanzhang/QuantLRM-R1-Qwen-32B-3-bit",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/nanzhang/QuantLRM-R1-Qwen-32B-3-bit

SGLang

How to use nanzhang/QuantLRM-R1-Qwen-32B-3-bit 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 "nanzhang/QuantLRM-R1-Qwen-32B-3-bit" \
    --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": "nanzhang/QuantLRM-R1-Qwen-32B-3-bit",
		"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 "nanzhang/QuantLRM-R1-Qwen-32B-3-bit" \
        --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": "nanzhang/QuantLRM-R1-Qwen-32B-3-bit",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use nanzhang/QuantLRM-R1-Qwen-32B-3-bit with Docker Model Runner:
```
docker model run hf.co/nanzhang/QuantLRM-R1-Qwen-32B-3-bit
```

Add metadata and improve model card

by nielsr HF Staff - opened Feb 10

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

-22

nielsr

Feb 10

Hi! I'm Niels from the Hugging Face community science team. I've opened this PR to enhance the model card with additional metadata and content improvements:

Pipeline Tag: Added text-generation to ensure the model is correctly categorized as a Large Language/Reasoning Model.
Library Name: Added transformers as the config.json indicates compatibility with the Transformers library, which enables the "Use in Transformers" code snippet.
Base Model: Explicitly linked to deepseek-ai/DeepSeek-R1-Distill-Qwen-32B to improve transparency and discoverability.
Additional Tags: Added reasoning to highlight the model's focus, and arxiv:2602.02581 to link it to its paper on the Hugging Face Papers page.
Content Structure: Streamlined the "Model Details" section for better readability and added the "Acknowledgement" section from the GitHub repository for completeness.
Cleanup: Removed the "Model Card Author" and "Model Card Contact" sections as they are typically not included in Hugging Face model cards.

These updates help users find and utilize your work more effectively. Let me know if you have any questions!

Add metadata and improve model carddbe28351

nanzhang changed pull request status to merged Feb 11

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