whereismyai
SWE-agent
SWE-agent is an open-source coding agent from Princeton and Stanford that uses a language model to fix GitHub issues.
SWE-agent lets a chosen language model, such as GPT-4o or Claude Sonnet 4, use tools on its own to fix issues in real GitHub repositories, look for security bugs, or run a custom task. A single YAML file governs the run. The authors report state-of-the-art SWE-bench results among open-source projects at several 2025 checkpoints, including a 32B open-weights model. Multimodal runs can read images from issues. The maintainers now recommend mini-SWE-agent, a much smaller successor with similar scores, and put most new work there. The project is built for research and is fully documented.
SWE-agent at a glance
- Best for: Software engineer
- Runs on: open source, CLI
- Pricing: Free and open source; model API usage is billed by the provider
- Released: April 2024
- Category: AI coding tool
- Website: swe-agent.com
Key features
- Autonomous tool use on real GitHub repositories
- Single YAML configuration
- SWE-bench results among open-source agents
- Multimodal issues with images
- Custom tasks including security review
- mini-SWE-agent successor in about 100 lines of Python
Use it for
- Reproduce a SWE-bench run for research
- Have a model open a pull request for a GitHub issue
- Compare mini-SWE-agent with the original agent
Companions
Tools that articles about SWE-agent recommend using with it, each with its job.
- Context7 — Provides up-to-date documentation for AI agents and code editors.
- Ollama — Run local models at zero marginal cost, with code kept on your machine.
- OpenRouter — Access over 500 models and 80 providers through one unified API.
- Daytona — Runs sessions in isolated sandboxes synced to git branches.
- Composio — Equips AI agents with tools, managed user auth, and sandboxes across 1,500+ apps
- Oh My OpenCode Slim — Lightweight multi-agent orchestration and background tasks.
Alternatives to SWE-agent
- Open SWE — It uses isolated cloud sandboxes and curated toolsets for production engineering.
- OpenCode — Works with 75+ model providers, local models, and existing subscriptions
- jevgrep — Reduces coding agent costs by providing targeted, relevant repository context.
- GitAgent — Manage AI agents with familiar git workflows like branching, diffing, and reverting.
- GoCodeo — Generates production-ready code, explains code, and creates tests in seconds.
- Adyen — Share AI agent resources across coding tools and repositories easily.
- Aider — Free terminal agent with a repo map and automatic git commits
- OpenClaude — Connects one terminal interface to over 200 different AI models.
- GitHub Agentic Workflows — Automates repository maintenance safely using AI agents in GitHub Actions.
- Flash Agents — Offloads reading and typing to cheap workers while keeping Claude for review
- Cline — Open source, any model, with approval on every step
- Agent QA — It uses repository-based memory to learn from test runs and perform self-healing.
All SWE-agent alternatives