Instructions to use ilgee/MetaMath-Mistral-7B-DFT2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ilgee/MetaMath-Mistral-7B-DFT2 with Transformers:

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

pipe = pipeline("text-generation", model="ilgee/MetaMath-Mistral-7B-DFT2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ilgee/MetaMath-Mistral-7B-DFT2")
model = AutoModelForCausalLM.from_pretrained("ilgee/MetaMath-Mistral-7B-DFT2")
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 ilgee/MetaMath-Mistral-7B-DFT2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ilgee/MetaMath-Mistral-7B-DFT2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ilgee/MetaMath-Mistral-7B-DFT2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ilgee/MetaMath-Mistral-7B-DFT2

SGLang

How to use ilgee/MetaMath-Mistral-7B-DFT2 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 "ilgee/MetaMath-Mistral-7B-DFT2" \
    --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": "ilgee/MetaMath-Mistral-7B-DFT2",
		"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 "ilgee/MetaMath-Mistral-7B-DFT2" \
        --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": "ilgee/MetaMath-Mistral-7B-DFT2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use ilgee/MetaMath-Mistral-7B-DFT2 with Docker Model Runner:
```
docker model run hf.co/ilgee/MetaMath-Mistral-7B-DFT2
```

Improve model card: Add pipeline tag, paper/code links, usage, and detailed info

by nielsr HF Staff - opened Jul 25, 2025

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

+213

-11

nielsr

Jul 25, 2025

This PR significantly improves the model card for siqi00/MetaMath-Mistral-7B-DFT2 by:

Adding the pipeline_tag: text-generation to the metadata, which ensures proper categorization and discoverability on the Hugging Face Hub.
Including direct links to the paper (Discriminative Finetuning of Generative Large Language Models without Reward Models and Human Preference Data) and the official GitHub repository (https://github.com/PenGuln/DFT).
Populating the "Model description", "Intended uses & limitations", and "Training and evaluation data" sections with detailed information extracted from the paper abstract and the project's GitHub README.
Adding comprehensive "Performance" tables for both mathematical reasoning and general language tasks, making the model's capabilities clear at a glance.
Providing a practical "Usage" example to help users quickly get started with text generation and chat completion.
Including detailed sections on "Installation", "Generating negative samples", "Evaluation", and "Precompute Log-likelihood" from the GitHub repository, enhancing reproducibility and practical utility.
Adding the BibTeX "Citation" for proper academic attribution.
Removing the automatically generated comment at the top.

This update makes the model card much more informative and user-friendly for the community.

Improve model card: Add pipeline tag, paper/code links, usage, and detailed info17934222

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

Ready to merge

This branch is ready to get merged automatically.

· Sign up or log in to comment