Git Stars Git Stars

#swe-bench Repositories

Discover 12 active open source GitHub repositories tagged with #swe-bench, accumulating 35,058 total stars (averaging 2,921 stars per project). Leading this topic is stitionai/devika with 19,560 stars.

12
Total Repositories
35,058
Total Stars
2,921
Avg Stars/Repo
1
stitionai/devika

Devika is the first open-source implementation of an Agentic Software Engineer. Initially started as an open-source alternative to Devin.

stitionai
โญ 19,560 Python ๐Ÿ‘ค stitionai Organization ๐Ÿ• Updated 2025-09-25 ๐Ÿ“… Created 2024-03-21
2
smallcloudai/refact

AI Agent that handles engineering tasks end-to-end: integrates with developersโ€™ tools, plans, executes, and iterates until it achieves a successful result.

smallcloudai
โญ 3,551 Rust ๐Ÿ‘ค smallcloudai Organization ๐Ÿ• Updated 2026-05-29 ๐Ÿ“… Created 2023-04-19
3
AutoCodeRoverSG/auto-code-rover

A project structure aware autonomous software engineer aiming for autonomous program improvement. Resolved 37.3% tasks (pass@1) in SWE-bench lite and 46.2% tasks (pass@1) in SWE-bench verified with each task costs less than $0.7.

AutoCodeRoverSG
โญ 3,105 Python ๐Ÿ‘ค AutoCodeRoverSG Organization ๐Ÿ• Updated 2026-11-01 ๐Ÿ“… Created 2024-04-08
4
jennyzzt/dgm

Darwin Gรถdel Machine: Open-Ended Evolution of Self-Improving Agents

jennyzzt
โญ 2,223 Python ๐Ÿ‘ค jennyzzt User ๐Ÿ• Updated 2025-08-13 ๐Ÿ“… Created 2025-05-23
5
MoonshotAI/Kimi-Dev

open-source coding LLM for software engineering tasks

MoonshotAI
โญ 1,357 Python ๐Ÿ‘ค MoonshotAI Organization ๐Ÿ• Updated 2025-09-30 ๐Ÿ“… Created 2025-06-16
6
facebookresearch/cwm

Research code artifacts for Code World Model (CWM) including inference tools, reproducibility, and documentation.

facebookresearch
โญ 922 Python ๐Ÿ‘ค facebookresearch Organization ๐Ÿ• Updated 2026-07-17 ๐Ÿ“… Created 2025-09-02
7
augmentcode/augment-swebench-agent

The #1 open-source SWE-bench Verified implementation

augmentcode
โญ 884 Python ๐Ÿ‘ค augmentcode Organization ๐Ÿ• Updated 2026-08-21 ๐Ÿ“… Created 2025-03-28
8
huggingface/Repo2RLEnv

Turn any repository into verifiable RL environments for coding agents - Harbor tasks you can train on, evaluate and share on the Hugging Face Hub

huggingface
โญ 804 Python ๐Ÿ‘ค huggingface Organization ๐Ÿ• Updated 2026-11-16 ๐Ÿ“… Created 2026-05-05
9
SWE-Gym/SWE-Gym

Code for Paper: Training Software Engineering Agents and Verifiers with SWE-Gym [ICML 2025]

SWE-Gym
โญ 764 Jupyter Notebook ๐Ÿ‘ค SWE-Gym Organization ๐Ÿ• Updated 2026-11-01 ๐Ÿ“… Created 2024-11-04
10
facebookresearch/swe-rl

[NeurIPS'25] Official codebase for "SWE-RL: Advancing LLM Reasoning via Reinforcement Learning on Open Software Evolution"

facebookresearch
โญ 696 Python ๐Ÿ‘ค facebookresearch Organization ๐Ÿ• Updated 2025-03-16 ๐Ÿ“… Created 2025-02-23
11
gersteinlab/LocAgent

[ACL 2025] Graph-guided agentic framework for code localization https://arxiv.org/abs/2503.09089

gersteinlab
โญ 642 Python ๐Ÿ‘ค gersteinlab Organization ๐Ÿ• Updated 2026-08-12 ๐Ÿ“… Created 2025-03-12
12
TIGER-AI-Lab/WebsiteBench

Can AI agents rebuild websites through browser exploration?

TIGER-AI-Lab
โญ 595 Python ๐Ÿ‘ค TIGER-AI-Lab Organization ๐Ÿ• Updated 2026-11-17 ๐Ÿ“… Created 2026-11-04

Frequently Asked Questions

1

What is the most popular #swe-bench repository on GitHub?

The top ranked open source #swe-bench project on GitHub is stitionai/devika with 19,560 stars. Devika is the first open-source implementation of an Agentic Software Engineer. Initially started as an open-source alternative to Devin.

2

Which programming languages are most used in #swe-bench repositories?

Among the top ranked #swe-bench projects on GitHub, the most common programming languages are Python, Rust, Jupyter Notebook.

3

How is the #swe-bench repository ranking calculated?

We track 12 active open source repositories tagged with #swe-bench (500+ stars), totaling 35,058 stars with an average of 2,921 stars per repository. Rankings update daily from the GitHub API.