Instructions to use entfane/math-genius-7B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use entfane/math-genius-7B with Transformers:

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

pipe = pipeline("text-generation", model="entfane/math-genius-7B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("entfane/math-genius-7B")
model = AutoModelForCausalLM.from_pretrained("entfane/math-genius-7B", 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=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use entfane/math-genius-7B with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/entfane/math-genius-7B

SGLang

How to use entfane/math-genius-7B 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 "entfane/math-genius-7B" \
    --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": "entfane/math-genius-7B",
		"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 "entfane/math-genius-7B" \
        --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": "entfane/math-genius-7B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use entfane/math-genius-7B with Docker Model Runner:
```
docker model run hf.co/entfane/math-genius-7B
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Math Genius 7B

This model is a Math Chain-of-Thought fine-tuned version of Mistral 7B v0.3 Instruct model.

Fine-tuning dataset

Model was fine-tuned on entfane/Mixture-Of-Thoughts-Math-No-COT math dataset.

Inference

!pip install transformers accelerate

from transformers import AutoTokenizer, AutoModelForCausalLM

model_name = "entfane/math-genius-7B"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)
messages = [
    {"role": "user", "content": "What's the derivative of 2x^2?"}
]
input = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
encoded_input = tokenizer(input, return_tensors = "pt").to(model.device)
output = model.generate(**encoded_input, max_new_tokens=1024)
print(tokenizer.decode(output[0], skip_special_tokens=False))

Evaluation

MathQA

The model was evaluated on a randomly sampled subset of 1,000 records from the test split of the Math-QA dataset. Math Genius 7B achieved an accuracy of 93.1% in producing the correct final answer under the pass@1 evaluation metric.