lliai/D2MoE
D^2-MoE: Delta Decompression for MoE-based LLMs Compression
This project helps machine learning engineers and researchers reduce the computational resources needed for large language models (LLMs) that use a Mixture-of-Experts (MoE) architecture. It takes an existing MoE LLM and outputs a compressed version that uses fewer parameters, making it faster and more memory-efficient to run, without needing to retrain the model. It's designed for those deploying or experimenting with large AI models.
No commits in the last 6 months.
Use this if you need to deploy or experiment with large MoE-based language models but are constrained by computational resources or want to improve inference speed.
Not ideal if you are looking to train a new LLM from scratch or if your models do not use the Mixture-of-Experts architecture.
Stars
74
Forks
8
Language
Python
License
Apache-2.0
Category
Last pushed
Mar 25, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/lliai/D2MoE"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
EfficientMoE/MoE-Infinity
PyTorch library for cost-effective, fast and easy serving of MoE models.
raymin0223/mixture_of_recursions
Mixture-of-Recursions: Learning Dynamic Recursive Depths for Adaptive Token-Level Computation...
AviSoori1x/makeMoE
From scratch implementation of a sparse mixture of experts language model inspired by Andrej...
thu-nics/MoA
[CoLM'25] The official implementation of the paper
jaisidhsingh/pytorch-mixtures
One-stop solutions for Mixture of Expert modules in PyTorch.