Kilian Lieret
4fce28706f
* Add ImageParsingHistoryProcessor * Add SWEBenchMultimodal problem statement parsing * cache processed swe-bench multimodal inputs * Add testing for multimodal problem statement * Update environment variable for PROBLEM_STATEMENT * Update timeout params to multimodal config * Add problem statement with images to Inspector view * Update configs and add default_mm_no_images * Add templates.disable_image_processing option * Add Observation Images to Inspector * Update configs for gemini * Update docs for multimodal updates * Fix: Protocols aren't inherited * Chore: pathlib instead of os.path and make pre-commit happy * add web_browser bundle --------- Co-authored-by: carlosejimenez <cjsaltlake@gmail.com> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
2.8 KiB
2.8 KiB
title
| title |
|---|
| Getting Started |
SWE-agent enables your language model of choice (e.g. GPT-4o or Claude Sonnet 4) to autonomously use tools to fix issues in real GitHub repositories, find cybersecurity vulnerabilities, or perform any custom task.
- ✅ State of the art on SWE-bench among open-source projects
- ✅ Free-flowing & generalizable: Leaves maximal agency to the LM
- ✅ Configurable & fully documented: Governed by a single
yamlfile - ✅ Made for research: Simple & hackable by design
SWE-agent is built and maintained by researchers from Princeton University and Stanford University.
-
:material-download:{ .lg .middle } Installation
Installing SWE-agent.
-
:material-cog:{ .lg .middle } Hello world
Solve a GitHub issue with SWE-agent.
-
:material-lightbulb:{ .lg .middle } User guides
Dive deeper into SWE-agent's features and goals.
-
:material-book:{ .lg .middle } Background & goals
Learn more about the project goals and academic research.
📣 News
- NEW: Multimodal support for SWE-bench - Process images from GitHub issues with vision-capable AI models
- May 2: SWE-agent-LM-32b achieves open-weights SOTA on SWE-bench
- Feb 28: SWE-agent 1.0 + Claude 3.7 is SoTA on SWE-Bench full
- Feb 25: SWE-agent 1.0 + Claude 3.7 is SoTA on SWE-bench verified
- Feb 13: Releasing SWE-agent 1.0: SoTA on SWE-bench light & tons of new features
- Dec 7: An interview with the SWE-agent & SWE-bench team
✍️ Doc updates
- June 26: Adding custom tools
- Apr 8: Running SWE-agent competitively
- Mar 7: Updated SWE-agent architecture diagram of 1.0