Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
[TKDE'25] The official GitHub page for the survey paper "A Survey on Mixture of Experts in Large Language Models".
| Date | Stars |
|---|---|
| 2026-07-31 | 505 |
| 2026-08-06 | 507 |
Today
+2 stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
<p align="center" width="100%">
<img src="assets/moe_logo_v2.jpeg" alt="HKUST MoE Survey" style="width: 25%; min-width: 250px; display: block; margin: auto;">
</p>
<div align="center">
<h1>A Survey on Mixture of Experts in<br>Large Language Models</h1>
<a href="https://awesome.re">
<img src="https://awesome.re/badge.svg" alt="Awesome">
</a>
<a href="https://img.shields.io/badge/PRs-Welcome-red">
<img src="https://img.shields.io/badge/PRs-Welcome-red" alt="PRs Welcome">
</a>
<a href="https://img.shields.io/github/last-commit/withinmiaov/A-Survey-on-Mixture-of-Experts?color=green">
<img src="https://img.shields.io/github/last-commit/withinmiaov/A-Survey-on-Mixture-of-Experts?color=green" alt="Last Commit">
</a>
</div>
<p align="center" width="100%">
<img src="assets/MoE_Timeline_20250723.jpg" alt="MoE LLMs Timeline" style="width: 95%; min-width: 100px; display: block; margin: auto;">
<br>
A chronological overview of several representative Mixture-of-Experts (MoE) models in recent years. The timeline is primarily structured according to the release dates of the models. MoE models located above the arrow are open-source, while those below the arrow are proprietary and closed-source. MoE models from various domains are marked with distinct colors: Natural Language Processing (NLP) in green, Computer Vision in yellow, Multimodal in pink, and Recommender Systems (RecSys) in cyan.
</p>
<p align="center" width="100%">
<img src="assets/MoE_Timeline_Jan22.jpg" alt="MoE LLMs Timeline" style="width: 95%; min-width: 100px; display: block; margin: auto;">
<br>
Previous Version: January 2025.
</p>
> [!IMPORTANT]
> **Good news! :tada: Our survey paper has been successfully accepted by TKDE. :fire::fire::fire:**
>
> A curated collection of papers and resources on Mixture of Experts in Large Language Models.
>
> Please refer to our survey [**"A Survey on Mixture of Experts in Large Language Models"**](https://arxiv.org/abs/2407.06204) for the detailed contents. [](https://arxiv.org/abs/2407.06204)
>
> Please let us know if you discover any mistakes or have suggestions by emailing us: [email protected]
## Table of Contents
- [Taxonomy](#taxonomy)
- [Paper List (Organized Chronologically and Categorically)](#paper-list-organized-chronologically-and-categorically)
- [Contributors](#contributors)
- [Star History](#star-history)
## Taxonomy
<p align="center" width="100%">
<img src="assets/moe_taxonomy.jpg" alt="MoE LLMs Taxonomy" style="width: 95%; min-width: 100px; display: block; margin: auto;">
</p>
<div align="right">
<b><a href="#table-of-contents">↥ back to top</a></b>
</div>
## Paper List (Organized Chronologically and Categorically)
- Less is MoE: Trimming Experts in Domain-Specialist Language Models, [[ArXiv 2026]](https://arxiv.org/abs/2606.05538), 2026-6-4
- LoopMoE: Unifying Iterative Computation with Mixture-of-Experts for Language Modeling, [[ArXiv 2026]](https://arxiv.org/abs/2606.04438), 2026-6-3
- UltraEP: Unleash MoE Training and Inference on Rack-Scale Nodes with Near-Optimal Load Balancing, [[ArXiv 2026]](https://arxiv.org/abs/2606.04101), 2026-6-2
- PRISM: Synergizing Vision Foundation Models via Self-organized Expert Specialization, [[ICML 2026]](https://arxiv.org/abs/2606.03444), 2026-6-2
- DOT-MoE: Differentiable Optimal Transport for MoEfication, [[ICML 2026]](https://arxiv.org/abs/2606.01666), 2026-6-1
- DAG-MoE: From Simple Mixture to Structural Aggregation in Mixture-of-Experts, [[ICML 2026]](https://arxiv.org/abs/2606.01062), 2026-5-31
- MESA: Improving MoE Safety Alignment via Decentralized Expertise, [[ICML 2026]](https://arxiv.org/abs/2606.00651), 2026-5-30
- How Far Can Disaggregation Go? A Design-Space Exploration of Attention-FFN Disaggregation for Efficient MoE LLM Serving, [[ArXiv 2026]](https://arxiv.org/abs/2605.28302), 2026-5-27
-Excerpt of 33,516 characters
Read on GitHub
62
16
11
2
Would you bet a product on this? Bounded 0–100 and slow moving.
matched fp:20470a1f991ae370, name:mixture of experts, desc:mixture of experts