benchmarks / arc-agi
ARC-AGI
score. Status active. Contamination risk med. Maintainer ARC Prize / Epoch collection. Source.
| Model | Score | Date | Kind |
|---|---|---|---|
| Claude Fable 5 | 0.985 | 2026-06-09 | observation |
| Claude Fable 5 | 0.985 | 2026-06-09 | observation |
| GPT-6 Astra | 0.985 | 2026-09-03 | observation |
| GPT-6 Astra | 0.985 | 2026-09-03 | observation |
| Claude Opus 5.5 | 0.985 | 2026-09-22 | observation |
| GPT-6.1 Sol | 0.985 | 2026-09-29 | observation |
| GPT-6.1 Sol | 0.985 | 2026-09-29 | observation |
| Gemini 3.8 Flash | 0.985 | 2026-09-02 | observation |
| Gemini 3.1 Pro Preview | 0.98 | 2026-02-19 | observation |
| GPT-5.6 Sol | 0.975 | 2026-07-09 | observation |
| Claude Opus 5 | 0.975 | 2026-07-24 | observation |
| Claude Opus 5 | 0.975 | 2026-07-24 | observation |
| Claude Fable 5.1 | 0.975 | 2026-09-01 | observation |
| GPT-6 Astra | 0.975 | 2026-09-03 | observation |
| GPT-6 Astra | 0.975 | 2026-09-03 | observation |
| Claude Opus 5.5 | 0.975 | 2026-09-22 | observation |
| Claude Opus 5.5 | 0.975 | 2026-09-22 | observation |
| Claude Opus 5.5 | 0.975 | 2026-09-22 | observation |
| Gemini 3.8 Flash | 0.975 | 2026-09-02 | observation |
| GPT-5.6 Sol | 0.97 | 2026-07-09 | observation |
| GPT-5.5 Pro | 0.965 | 2026-04-23 | observation |
| GPT-5.6 Sol | 0.965 | 2026-07-09 | observation |
| GPT-5.6 Terra | 0.965 | 2026-07-09 | observation |
| Claude Fable 5.1 | 0.965 | 2026-09-01 | observation |
| GPT-6 Astra | 0.965 | 2026-09-03 | observation |
| GPT-6.1 Sol | 0.965 | 2026-09-29 | observation |
| Gemini 3 Deep Think Preview | 0.96 | 2026-02-12 | observation |
| Claude Fable 5.1 | 0.96 | 2026-09-01 | observation |
| Claude Fable 5 | 0.9550000000000001 | 2026-06-09 | observation |
| Gemini 3.7 Flash | 0.9550000000000001 | 2026-08-13 | observation |