Coding Agent Leaderboard - SWE-Bench Verified (Claude Code): leaderboard
Red Hat's Coding Agent Leaderboard run of SWE-bench Verified — the 500 human-validated GitHub issues from Python repositories, with the Claude Code agent scaffold in Harbor sandboxes; percentage of tasks resolved.
Metric: Score (%). Source: huggingface.co. Saturation forecast: Around February 2027. 10 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.8 | 86.8 | #56 |
| 2 | Qwen 3.8 27B | 86.8 | #204 |
| 3 | Claude Sonnet 4.6 | 79.6 | #161 |
| 4 | GPT-OSS-120B | 37.6 | #700 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=coding-agent-leaderboard-swe-bench-verified-claude-code · How It Works · Data refreshed daily, snapshot 2026-11-13.