Instructions to use TencentARC/MetaMath-Mistral-Pro with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use TencentARC/MetaMath-Mistral-Pro with Transformers:

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

pipe = pipeline("text-generation", model="TencentARC/MetaMath-Mistral-Pro")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("TencentARC/MetaMath-Mistral-Pro")
model = AutoModelForCausalLM.from_pretrained("TencentARC/MetaMath-Mistral-Pro")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use TencentARC/MetaMath-Mistral-Pro with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "TencentARC/MetaMath-Mistral-Pro"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "TencentARC/MetaMath-Mistral-Pro",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/TencentARC/MetaMath-Mistral-Pro

SGLang

How to use TencentARC/MetaMath-Mistral-Pro 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 "TencentARC/MetaMath-Mistral-Pro" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "TencentARC/MetaMath-Mistral-Pro",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "TencentARC/MetaMath-Mistral-Pro" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "TencentARC/MetaMath-Mistral-Pro",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use TencentARC/MetaMath-Mistral-Pro with Docker Model Runner:
```
docker model run hf.co/TencentARC/MetaMath-Mistral-Pro
```

WuChengyue commited on Feb 27, 2024

Commit

3835d38

verified ·

1 Parent(s): bd9406c

Update README.md

Browse files

Files changed (1) hide show

README.md +7 -9

README.md CHANGED Viewed

@@ -21,17 +21,15 @@ MetaMath-Mistral-Pro is fully fine-tuned on the MetaMathQA datasets and based on
 ## Model Usage
-prompting template:
-'''
-"Below is an instruction that describes a task. "
-"Write a response that appropriately completes the request.\n\n"
-"### Instruction:\n{instruction}\n\n### Response: Let's think step by step."
-'''
-where you need to use your query question to replace the {instruction}
 ## Experiments

 ## Model Usage
+The model is trained to use the following format (note the newlines):
+```
+<|user|>
+Your message here!
+<|assistant|>
+```
+For best results, format all inputs in this manner. **Make sure to include a newline after `<|assistant|>`, this can affect generation quality quite a bit.**
 ## Experiments