Optimal Sparsity Code
Collection
Optimal Sparsity of Mixture-of-Experts Language Models for Reasoning Tasks • 65 items • Updated • 1
Text Generation
Transformers
Safetensors
mixtral
Mixture of Experts
reasoning
text-generation-inference
How to use from
vLLMUse Docker
docker model run hf.co/llm-jp/optimal-sparsity-code-d512-E128-k4-3.3B-A220MQuick Links
Optimal Sparsity of Mixture-of-Experts Language Models for Reasoning Tasks
This repository contains model checkpoints from the paper Optimal Sparsity of Mixture-of-Experts Language Models for Reasoning Tasks.
For more details, including code and evaluation procedures, please refer to the official GitHub repository: https://github.com/rioyokotalab/optimal-sparsity
How to cite
If you find our work helpful, please feel free to cite the paper.
@inproceedings{
nakamura2026optimal,
title={Optimal Sparsity of Mixture-of-Experts Language Models for Reasoning Tasks},
author={Taishi Nakamura and Satoki Ishikawa and Masaki Kawamura and Takumi Okamoto and Daisuke Nohara and Jun Suzuki and Rio Yokota},
booktitle={The Fourteenth International Conference on Learning Representations},
year={2026},
url={https://openreview.net/forum?id=XFw2EPRUUR}
}
- Downloads last month
- 33
Install from pip and serve model
# Install vLLM from pip: pip install vllm# Start the vLLM server: vllm serve "llm-jp/optimal-sparsity-code-d512-E128-k4-3.3B-A220M"# Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "llm-jp/optimal-sparsity-code-d512-E128-k4-3.3B-A220M", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'