SWE Atlas (All Workflows): leaderboard

SWE Atlas (Scale AI, 2026) evaluates coding agents on 284 tasks in real repositories across three professional workflows, Codebase Q and A, Test Writing and Refactoring, scored by programmatic checks plus must-have rubrics, with three trials per task and each model run in an agent harness. This board is the pass rate in percent averaged over trials and all three workflows; higher is better.

Metric: Pass@1 (%): mean per-trial pass rate over all 284 SWE Atlas tasks (124 Codebase Q&A, 90 Test Writing and 70 Refactoring tasks in real repositories, scored by programmatic checks plus must-have rubrics), three trials per task run through the Harbor framework, each model in its provider's own agent harness (Codex CLI, Claude Code, Gemini CLI) or in mini-SWE-agent; higher is better. Source: arxiv.org. Saturation forecast: Around March 2028. 9 models tracked. Updated 7 Oct 2026.

Top models

#ModelSettingScoreOverall rank
1Claude Opus 4.7xHigh38.94#85
2GPT-5.4xHigh38#135
3Claude Opus 4.6High31.83#113
4Claude Sonnet 4.6High31.63#173
5GLM-524.03#269
6Gemini 3.1 Pro (Preview)High23.73#99
7Kimi K2.519.05#266
8Gemini 3 FlashHigh15.65#175
9MiniMax-M2.515.2#600

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=swe-atlas-all-workflows · How It Works · Data refreshed daily, snapshot 2026-11-21.