Benchmarks / External Composite Indices
AA Coding Index
Artificial Analysis's coding composite. Displayed for reference only.
maintained by Artificial Analysis · 94 observations on record
Versions
1 version
| Version | Tasks |
|---|---|
| AA Coding Index current aa-coding-index | — |
Current leaderboard
AA Coding Index · Index points · top 25 of 94 systems
| # | System | Index points |
|---|---|---|
| 1 | GPT 5.6 Solxhigh effort OpenAI | 78.3 points |
| 2 | GPT 5.6 Solmax effort OpenAI | 77.4 points |
| 3 | GPT 5.6 Solhigh effort OpenAI | 77.2 points |
| 4 | GPT 5.6 Terramax effort OpenAI | 76.7 points |
| 5 | GPT 5.6 Solmedium effort OpenAI | 76.3 points |
| 6 | GPT 5.5 (2026-04-22)xhigh effort OpenAI | 74.9 points |
| 7 | Claude Opus 4.8max effort Anthropic | 74.3 points |
| 8 | Claude Opus 4.7max effort Anthropic | 73.6 points |
| 9 | Grok 4.5high effort xAI | 72.4 points |
| 10 | GPT 5.5 (2026-04-22)high effort OpenAI | 71.6 points |
| 11 | GPT 5.5 (2026-04-22)medium effort OpenAI | 71.5 points |
| 12 | Claude Sonnet 5max effort Anthropic | 71.5 points |
| 13 | GPT 5.6 Lunamax effort OpenAI | 71.4 points |
| 14 | Muse Spark 1.1xhigh effort Meta AI | 71.3 points |
| 15 | GPT 5.4 (2026-03-05)xhigh effort OpenAI | 71.1 points |
| 16 | GPT 5.6 Terraxhigh effort OpenAI | 70.6 points |
| 17 | Gemini 3.5 Flashhigh effort Google DeepMind | 70.1 points |
| 18 | GPT 5.6 Sollow effort OpenAI | 69.7 points |
| 19 | Gemini 3.1 Pro Preview Google DeepMind | 68.8 points |
| 20 | GLM 5.2max effort Z.ai | 68.8 points |
| 21 | GPT 5.6 Lunaxhigh effort OpenAI | 68.6 points |
| 22 | GPT 5.6 Terrahigh effort OpenAI | 67.1 points |
| 23 | Claude Sonnet 5high effort Anthropic | 66.4 points |
| 24 | Qwen3 7 Max Alibaba (Qwen) | 66.0 points |
| 25 | GPT 5.6 Solno reasoning OpenAI | 65.1 points |
One row per evaluated system — reasoning-effort variants rank separately.