The winners code of the LLM Merging Competition at NeurIPS 2025
This repository hosts the code submissions from the winning teams of the LLM-Merging Competition at NeurIPS 2025. Congratulations to all winners for their outstanding contributions!
We have five winning submissions across three tracks. Each submission’s code is provided as a .tar file, named as follows:
The code are hosted on Google drive
| Placement | Paper & Authors |
|---|---|
| 1st Place 🌟 | A Model Merging Method Qiang Gao, Jisheng Fang, Hao Mo |
| 2nd Place ⭐ | Simple Llama Merge: What Kind of LLM Do We Need? Yinuo Zhang |
| 3rd Place 💫 | LLM Merging Competition Technical Report: Efficient Model Merging with Strategic Model Selection, Merging, and Hyperparameter Optimization Zixiang Di, Yaoming Yang, Mei Jiang, Bingdong Li, Hong Qian, Aimin Zhou |
| Winner 🚀 |
|---|
| LLM Merging Competition Technical Report for NeurIPS 2024: Efficiently Building Large Language Models through Merging Yizhen Zhang, Yang Ding, Jie Wu, Yujiu Yang |
| Winner 🎆 |
|---|
| Model Merging using Geometric Median of Task Vectors Siddharth Gupta, Aakash Gupta |