Instructions to use HangGuo/QWen2.5-7B-FlatQuant-OBR-GPTQ-W4A4KV4S50 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use HangGuo/QWen2.5-7B-FlatQuant-OBR-GPTQ-W4A4KV4S50 with Transformers:

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

pipe = pipeline("text-generation", model="HangGuo/QWen2.5-7B-FlatQuant-OBR-GPTQ-W4A4KV4S50")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("HangGuo/QWen2.5-7B-FlatQuant-OBR-GPTQ-W4A4KV4S50")
model = AutoModelForCausalLM.from_pretrained("HangGuo/QWen2.5-7B-FlatQuant-OBR-GPTQ-W4A4KV4S50")
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 HangGuo/QWen2.5-7B-FlatQuant-OBR-GPTQ-W4A4KV4S50 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "HangGuo/QWen2.5-7B-FlatQuant-OBR-GPTQ-W4A4KV4S50"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "HangGuo/QWen2.5-7B-FlatQuant-OBR-GPTQ-W4A4KV4S50",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/HangGuo/QWen2.5-7B-FlatQuant-OBR-GPTQ-W4A4KV4S50

SGLang

How to use HangGuo/QWen2.5-7B-FlatQuant-OBR-GPTQ-W4A4KV4S50 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 "HangGuo/QWen2.5-7B-FlatQuant-OBR-GPTQ-W4A4KV4S50" \
    --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": "HangGuo/QWen2.5-7B-FlatQuant-OBR-GPTQ-W4A4KV4S50",
		"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 "HangGuo/QWen2.5-7B-FlatQuant-OBR-GPTQ-W4A4KV4S50" \
        --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": "HangGuo/QWen2.5-7B-FlatQuant-OBR-GPTQ-W4A4KV4S50",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use HangGuo/QWen2.5-7B-FlatQuant-OBR-GPTQ-W4A4KV4S50 with Docker Model Runner:
```
docker model run hf.co/HangGuo/QWen2.5-7B-FlatQuant-OBR-GPTQ-W4A4KV4S50
```

Improve model card: Add OBR paper details, usage, and tags

by nielsr HF Staff - opened Sep 18, 2025

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

+89

-135

nielsr

Sep 18, 2025

This PR significantly improves the model card for the Qwen2.5-7B-FlatQuant-OBR-GPTQ-W4A4KV4S50 model (as implied by the context of FlatQuant and Qwen2.5).

Key changes include:

Replacing the boilerplate content with a detailed description, including the paper title and abstract.
Adding a link to the official Hugging Face paper page: Optimal Brain Restoration for Joint Quantization and Sparsification of LLMs.
Adding a link to the GitHub repository: https://github.com/csguoh/OBR.
Setting the pipeline_tag to text-generation to improve discoverability.
Updating the license to mit, aligning with the underlying FlatQuant method used for this specific model.
Adding relevant tags: quantization, sparsity, llm, qwen2.
Including a clear "How to Get Started" section with a functional code snippet for transformers inference, adapted from best practices and the context provided in the GitHub repository.
Adding the BibTeX citation.
Populating various "Model Details" fields and removing empty boilerplate sections.

These changes make the model card much more informative and user-friendly, aligning it with Hugging Face's best practices.

Improve model card: Add OBR paper details, usage, and tags35948b1c

HangGuo changed pull request status to merged Sep 26, 2025

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