Instructions to use LLM360/MegaMath-Llama-3.2-1B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use LLM360/MegaMath-Llama-3.2-1B with Transformers:

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

pipe = pipeline("text-generation", model="LLM360/MegaMath-Llama-3.2-1B")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("LLM360/MegaMath-Llama-3.2-1B")
model = AutoModelForCausalLM.from_pretrained("LLM360/MegaMath-Llama-3.2-1B")

Inference
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use LLM360/MegaMath-Llama-3.2-1B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "LLM360/MegaMath-Llama-3.2-1B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "LLM360/MegaMath-Llama-3.2-1B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/LLM360/MegaMath-Llama-3.2-1B

SGLang

How to use LLM360/MegaMath-Llama-3.2-1B 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 "LLM360/MegaMath-Llama-3.2-1B" \
    --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": "LLM360/MegaMath-Llama-3.2-1B",
		"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 "LLM360/MegaMath-Llama-3.2-1B" \
        --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": "LLM360/MegaMath-Llama-3.2-1B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use LLM360/MegaMath-Llama-3.2-1B with Docker Model Runner:
```
docker model run hf.co/LLM360/MegaMath-Llama-3.2-1B
```

Add link to code repository

by nielsr HF Staff - opened Apr 9, 2025

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

-4

Files changed (1) hide show

README.md +4 -4

README.md CHANGED Viewed

@@ -1,11 +1,11 @@
 ---
-license: llama3.2
 datasets:
 - LLM360/MegaMath
 language:
 - en
-pipeline_tag: text-generation
 library_name: transformers
 tags:
 - math
 - code
@@ -15,7 +15,7 @@ tags:
 # MegaMath-Llama-3.2-1B
-[Arxiv](arxiv.org/abs/2504.02807) | [Datasets](https://huggingface.co/datasets/LLM360/MegaMath)
 A proof-of-concept model train on [MegaMath](https://huggingface.co/datasets/LLM360/MegaMath) dataset, capable of both Chain-of-Thought and Program-Aided-Language problem solving.
@@ -32,7 +32,7 @@ If you find our work useful, please cite
 @article{zhou2025megamath,
   title     = {MegaMath: Pushing the Limits of Open Math Corpora},
   author    = {Zhou, Fan and Wang, Zengzhi and Ranjan, Nikhil and Cheng, Zhoujun and Tang, Liping and He, Guowei and Liu, Zhengzhong and Xing, Eric P.},
-  journal   = {arXiv preprint arXiv:2504.xxxxx},
   year      = {2025},
   note      = {Preprint}
 }

 ---
 datasets:
 - LLM360/MegaMath
 language:
 - en
 library_name: transformers
+license: llama3.2
+pipeline_tag: text-generation
 tags:
 - math
 - code
 # MegaMath-Llama-3.2-1B
+[Arxiv](arxiv.org/abs/2504.02807) | [Datasets](https://huggingface.co/datasets/LLM360/MegaMath) | [Code](https://github.com/LTH14/mar)
 A proof-of-concept model train on [MegaMath](https://huggingface.co/datasets/LLM360/MegaMath) dataset, capable of both Chain-of-Thought and Program-Aided-Language problem solving.
 @article{zhou2025megamath,
   title     = {MegaMath: Pushing the Limits of Open Math Corpora},
   author    = {Zhou, Fan and Wang, Zengzhi and Ranjan, Nikhil and Cheng, Zhoujun and Tang, Liping and He, Guowei and Liu, Zhengzhong and Xing, Eric P.},
+  journal   = {arXiv preprint arXiv:2504.02807},
   year      = {2025},
   note      = {Preprint}
 }