AA Coding Agent Index v2.0: leaderboard
Artificial Analysis Coding Agent Index v2.0 (October 2026): each row is a model running inside a coding-agent harness such as Claude Code, Codex or Mistral Vibe, scored as the equal-weight average of FrontierCode v1.1 Main, Terminal-Bench 4.0 and SWE-Atlas-QnA-AA over three attempts per task. Unlike AA's model-level evaluations, the coding agent named in each row is part of what is measured; v2.0 replaced DeepSWE with FrontierCode, so earlier Coding Agent Index figures are not comparable. Router and fusion agents, and cells where refusals stopped more than 5% of attempts, are excluded. Index (0-100).
Metric: Coding Agent Index (0-100). Source: artificialanalysis.ai. Saturation forecast: Around 2028. 30 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 4 Carbon (High) | 59.4 | #1 (Gemini 4 Carbon) |
| 2 | Muse Wonder Full (Max) | 58.88 | #12 (Muse Wonder Full) |
| 3 | GPT-6 Astra (Max) | 54.66 | #6 (GPT-6 Astra) |
| 4 | GPT-6.1 Sol (Max) | 54.13 | #7 (GPT-6.1 Sol) |
| 5 | Gemini 4 Argon (High) | 53.25 | #9 (Gemini 4 Argon) |
| 6 | Muse Wonder (Max) | 51.98 | #18 (Muse Wonder) |
| 7 | GLM-5.3 (Max) | 48.92 | #44 (GLM-5.3) |
| 8 | mistral-large-4-oss (High) | 48.28 | #60 (mistral-large-4-oss) |
| 9 | GPT-6.1 Terra (xHigh) | 48.22 | #15 (GPT-6.1 Terra) |
| 10 | GPT-6.1 Terra (Max) | 48.1 | #15 (GPT-6.1 Terra) |
| 11 | DeepSeek V4.1 Flash (Max) | 47.3 | #37 (DeepSeek V4.1 Flash) |
| 12 | Grok 4.7 (xHigh) | 46.6 | #26 (Grok 4.7) |
| 13 | DeepSeek V4.1 Flash (1017) (Max) | 46.31 | #27 (DeepSeek V4.1 Flash (1017)) |
| 14 | GPT-6.1 Terra (High) | 46.28 | #15 (GPT-6.1 Terra) |
| 15 | Muse Spark 1.3 (Max) | 45.71 | #31 (Muse Spark 1.3) |
No result here: #3 Claude Opus 5.5, #10 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=aa-coding-agent-index-v2-0 · How It Works · Data refreshed daily, snapshot 2026-11-05.