AltHub

SWE-agent/SWE-agent Alternatives

autonomous coding agents / SWE-bench

SWE-agent is an LLM-powered autonomous agent that uses tools to fix GitHub issues and perform coding tasks, with configurable agent loops.

20.2kPythonMITactive · last push 2mo agoGitHub

Drop-in peers

Same problem, same approach. Swapping one for another is a config change, not a rewrite.

augmentcode/augment-swebench-agentGitHub

choose if you want a focused SWE-bench Verified harness rather than SWE-agent's broader research framework.

The #1 open-source SWE-bench Verified implementation

882
Python
NOASSERTION
active2mo ago

Same job, different approach

Solves the same problem with a different architecture or at a different layer. Expect to rewrite the integration.

earendil-works/piGitHub

choose if you want a TypeScript/TUI-based agent toolkit instead of a Python harness.

AI agent toolkit: unified LLM API, agent loop, TUI, coding agent CLI

97.8k
TypeScript
MIT
active2mo ago
esengine/DeepSeek-ReasonixGitHub

choose if you want a Go terminal agent optimized for DeepSeek and prefix caching.

DeepSeek-native AI coding agent for your terminal. Engineered around prefix-cache stability — leave it running.

35.5k
Go
MIT
active1mo ago
AutoCodeRoverSG/auto-code-roverGitHub

choose if you prefer structure-aware program repair over SWE-agent's generic tool loop.

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.

3.1k
Python
NOASSERTION
slowing1.5y ago
china-qijizhifeng/agentic-harness-engineeringGitHub

choose if you are targeting Terminal-Bench instead of SWE-bench with an evolved harness.

Official AHE code — Agentic Harness Engineering: observability-driven automatic evolution of coding-agent harnesses (concurrent w/ meta-harness). NexAU-AHE reaches 84.7% ± 2.1 pass@1 on Terminal-Bench 2 (GPT-5.5). Lifts GPT-5.4 69.7→77.0% over 10 iters, beats Codex/ACE/Training-Free GRPO; frozen harness transfers to SWE-bench-Verified.

860
Python
MIT
active2mo ago
Danau5tin/multi-agent-coding-systemGitHub

choose if you want a multi-agent orchestrator for terminal coding tasks rather than SWE-agent's single-agent loop.

Reached #13 on Stanford's Terminal Bench leaderboard. Orchestrator, explorer & coder agents working together with intelligent context sharing.

1.4k
Python
Apache-2.0
slowing0.9y ago

Listed as an alternative to

These projects were analysed and named SWE-agent among their alternatives. The relationship is not symmetric — how SWE-agent rates them is a separate judgement, made when SWE-agent is analysed in its own right.

langchain-ai/open-sweGitHub

calls SWE-agent “Choose when you want a proven GitHub-issue-fixing agent with pluggable LMs instead of building on LangGraph.”

An Open-Source Asynchronous Coding Agent

10.6kPythonactive
FoundationAgents/MetaGPTGitHub

calls SWE-agent “Narrower issue-fixing agent; choose if your task is fixing GitHub issues rather than generating a full project from spec.”

🌟 The Multi-Agent Framework: First AI Software Company, Towards Natural Language Programming

70.1kPythonslowing