Instructions to use AI-MO/NuminaMath-7B-TIR-GPTQ with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use AI-MO/NuminaMath-7B-TIR-GPTQ with Transformers:

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

pipe = pipeline("text-generation", model="AI-MO/NuminaMath-7B-TIR-GPTQ")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("AI-MO/NuminaMath-7B-TIR-GPTQ")
model = AutoModelForCausalLM.from_pretrained("AI-MO/NuminaMath-7B-TIR-GPTQ", 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 AI-MO/NuminaMath-7B-TIR-GPTQ with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/AI-MO/NuminaMath-7B-TIR-GPTQ

SGLang

How to use AI-MO/NuminaMath-7B-TIR-GPTQ 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 "AI-MO/NuminaMath-7B-TIR-GPTQ" \
    --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": "AI-MO/NuminaMath-7B-TIR-GPTQ",
		"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 "AI-MO/NuminaMath-7B-TIR-GPTQ" \
        --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": "AI-MO/NuminaMath-7B-TIR-GPTQ",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use AI-MO/NuminaMath-7B-TIR-GPTQ with Docker Model Runner:
```
docker model run hf.co/AI-MO/NuminaMath-7B-TIR-GPTQ
```

Model Card for NuminaMath 7B TIR GPTQ

NuminaMath is a series of language models that are trained to solve math problems using tool-integrated reasoning (TIR). NuminaMath 7B TIR won the first progress prize of the AI Math Olympiad (AIMO), with a score of 29/50 on the public and private tests sets.

This model is an 8-bit version of AI-MO/NuminaMath-7B-TIR, which we quantized with AutoGPTQ to run fast inference in the Kaggle submissions. Please consult the original model card for more details.

Downloads last month: 10

Model tree for AI-MO/NuminaMath-7B-TIR-GPTQ

Base model

deepseek-ai/deepseek-math-7b-base

Quantized

(8)

this model

Collection including AI-MO/NuminaMath-7B-TIR-GPTQ

AIMO Progress Prize

Collection

Models and datasets used in the winning solution to the AIMO 1st Progress Prize • 7 items • Updated Jul 19, 2024 • 14