Instructions to use TMLR-Group-HF/Entropy-Qwen3-4B-Base-MATH with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use TMLR-Group-HF/Entropy-Qwen3-4B-Base-MATH with Transformers:

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

pipe = pipeline("text-generation", model="TMLR-Group-HF/Entropy-Qwen3-4B-Base-MATH")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("TMLR-Group-HF/Entropy-Qwen3-4B-Base-MATH")
model = AutoModelForCausalLM.from_pretrained("TMLR-Group-HF/Entropy-Qwen3-4B-Base-MATH", device_map="auto")
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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use TMLR-Group-HF/Entropy-Qwen3-4B-Base-MATH with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "TMLR-Group-HF/Entropy-Qwen3-4B-Base-MATH"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "TMLR-Group-HF/Entropy-Qwen3-4B-Base-MATH",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/TMLR-Group-HF/Entropy-Qwen3-4B-Base-MATH

SGLang

How to use TMLR-Group-HF/Entropy-Qwen3-4B-Base-MATH 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 "TMLR-Group-HF/Entropy-Qwen3-4B-Base-MATH" \
    --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": "TMLR-Group-HF/Entropy-Qwen3-4B-Base-MATH",
		"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 "TMLR-Group-HF/Entropy-Qwen3-4B-Base-MATH" \
        --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": "TMLR-Group-HF/Entropy-Qwen3-4B-Base-MATH",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use TMLR-Group-HF/Entropy-Qwen3-4B-Base-MATH with Docker Model Runner:
```
docker model run hf.co/TMLR-Group-HF/Entropy-Qwen3-4B-Base-MATH
```

Improve model card: Add pipeline tag, library name, paper link, abstract, and update GitHub/citation

by nielsr HF Staff - opened Oct 11, 2025

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

+16

-10

nielsr

Oct 11, 2025

This PR significantly improves the model card for TMLR-Group-HF/Entropy-Qwen3-4B-Base by:

Adding pipeline_tag: text-generation to ensure it is discoverable under text generation models.
Adding library_name: transformers to enable the "Use in Transformers" widget and provide a clear usage indicator.
Linking directly to the Hugging Face paper: Co-rewarding: Stable Self-supervised RL for Eliciting Reasoning in Large Language Models.
Including the full paper abstract to provide comprehensive context on the model's background and methodology.
Updating the GitHub repository link to the correct casing and formatting it as a proper Markdown link.
Updating the BibTeX citation to match the one provided in the official GitHub repository.

These updates will make the model more accessible, discoverable, and informative for researchers and developers on the Hugging Face Hub.

Improve model card: Add pipeline tag, library name, paper link, abstract, and update GitHub/citation6361c8c5

Geraldxm changed pull request status to merged Oct 11, 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