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.
SE-Agent is a self-evolution framework for LLM Code agents. It enables trajectory-level evolution to exchange information across reasoning paths via Revision, Recombination, and Refinement, expanding the search space and escaping local optima. On SWE-bench Verified, it achieves SOTA performance
| Date | Stars |
|---|---|
| 2026-07-31 | 281 |
| 2026-08-06 | 281 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
<div align="center">
<table border="0">
<tr>
<td><img src="static/img/logo.png" alt="SE-Agent Logo" width="250"/></td>
<td><h1>SE-Agent: Self-Evolution Trajectory Optimization in Multi-Step Reasoning with LLM-Based Agents</h1></td>
</tr>
</table>
</div>
<p align="center">
<strong>🏆 State-of-the-Art Performance on SWE-bench Verified: 80%</strong><br>
</p>
<p align="center">
<a href="https://arxiv.org/abs/2508.02085"><img src="https://img.shields.io/badge/Paper-arXiv-red.svg" alt="Paper"></a>
<a href="https://quantaalpha.github.io/"><img src="https://img.shields.io/badge/Homepage-QuantaAlpha-00A98F.svg?style=flat-square&logo=opensourceinitiative&logoColor=white" /></a>
<a href="https://www.swebench.com"><img src="https://img.shields.io/badge/SWE--bench-Verified%20-gold.svg" alt="SWE-bench"></a>
<a href="LICENSE"><img src="https://img.shields.io/badge/License-MIT-green.svg" alt="License"></a>
</p>
## 📰 News
- **2025.09.19** 🎉 Excited to announce that our papers have been accepted to <u>**NeurIPS 2025**</u> — [**RepoMaster**](https://arxiv.org/abs/2505.21577) as a ***Spotlight*** (≈3.2%) and [**SE-Agent**](https://arxiv.org/abs/2508.02085) as a ***Poster*** (≈24.52%)!
- **2025.08.28** 🎉 We open-sourced [**RepoMaster**](https://github.com/QuantaAlpha/RepoMaster) — an AI agent that leverages GitHub repos to solve complex real-world tasks.
- **2025.08.26** 🎉 We open-sourced [**GitTaskBench**](https://github.com/QuantaAlpha/GitTaskBench) — a repo-level benchmark & tooling suite for real-world tasks.
- **2025.08.10** 🎉 We open-sourced [**SE-Agent**](https://github.com/JARVIS-Xs/SE-Agent) — a self-evolution trajectory framework for multi-step reasoning.
> 🔗 **Ecosystem**: [RepoMaster](https://github.com/QuantaAlpha/RepoMaster) · [GitTaskBench](https://github.com/QuantaAlpha/GitTaskBench) · [SE-Agent](https://github.com/JARVIS-Xs/SE-Agent) · [Team Homepage](https://quantaalpha.github.io)
---
## 🎯 What is SE-Agent?
**SE-Agent** is a **self-evolution framework** that enables information exchange between reasoning paths through a trajectory-level evolution mechanism, breaking the cognitive limitations of single trajectories. This approach expands the search space, escapes local optima, and demonstrates emergent capabilities from collective interactions, achieving **80% Top1 performance** on **SWE-bench Verified**, truly realizing autonomous evolution for LLM-based agents in complex reasoning tasks.
<div align="center">
<img src="static/img/framework.jpg" alt="SE-Agent Framework" width="800px" />
</div>
### 📊 Performance Results
#### 🏆 SWE-bench Verified Evaluation: State-of-the-Art Performance!
SE-Agent ranks **Top1 among open-source frameworks** on SWE-bench Verified.
<div align="center">
<img src="static/img/ranking_new.jpg" alt="Open-Source Framework Comparison" width="800px" />
</div>
#### ✨ Performance Comparison: Leading with Significant Gains!
SE-Agent gain significant gains on SWE-bench Verified **across various open-source/closed-source LLMs**.
<div align="center">
<img src="static/img/main_result_table.png" alt="SE-Agent Framework Illustration" width="800px" />
</div>
## ⚡ Quick Start
Get SE-Agent running in **30 seconds**:
```bash
# 1. Clone and install
git clone https://github.com/JARVIS-Xs/SE-Agent.git
cd SE-Agent
pip install -e .
# 2. Set up API key
echo "DEEPSEEK_API_KEY=your_key_here" > .env
# 3. Run demo (no API calls)
python SE/basic_run.py --mode demo
# 4. Run your first experiment
python SE/basic_run.py --mode execute
```
**Expected Output:**
```text
✅ SE-Agent initialized successfully
🔄 Starting self-evolution with 3 iterations
```
> 💡 **Need detailed setup?** See [📦 Installation & Configuration](#-installation--configuration) below
## 🧠 How SE-Agent Works
SE-Agent implements **three core self-evolution operations** that transform how agents approach problem-solving:
### 🔄 Three Core Operations
#### 1. **🔧 Revision** - *Failure-Driven Strategy GeneratioExcerpt of 12,260 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:73fba443d15a80f9, llm:Repository description: 'SE-Agent is a self-evolution framework for LLM Code agents... enables trajectory-level evolution... Revision, Recombination, and Refinement... On SWE-bench Verified, it achieves SOTA performance'. Topics: claude-code, code-agent, code-fix, mcts, self-evolve, swe-agent, swe-bench, test-time-scaling.
matched fp:73fba443d15a80f9, llm:Repository description: 'SE-Agent is a self-evolution framework for LLM Code agents... enables trajectory-level evolution... Revision, Recombination, and Refinement... On SWE-bench Verified, it achieves SOTA performance'. Topics: claude-code, code-agent, code-fix, mcts, self-evolve, swe-agent, swe-bench, test-time-scaling.
matched fp:73fba443d15a80f9, llm:Repository description: 'SE-Agent is a self-evolution framework for LLM Code agents... enables trajectory-level evolution... Revision, Recombination, and Refinement... On SWE-bench Verified, it achieves SOTA performance'. Topics: claude-code, code-agent, code-fix, mcts, self-evolve, swe-agent, swe-bench, test-time-scaling.