IOI: leaderboard
Based on the International Olympiad in Informatics.
Metric: Accuracy (%). Source: vals.ai. Saturation forecast: Around August 2027. 44 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6 Astra | 100 | #6 |
| 2 | Gemini 4 Argon | 100 | #8 |
| 3 | GPT-6.1 Sol | 96.89 | #7 |
| 4 | Claude Opus 5.5 | 95.06 | #3 |
| 5 | Best Expert | 93.86 | |
| 6 | GPT-5.6 Sol | 91.17 | #19 |
| 7 | Claude Fable 5.1 | 90.78 | #9 |
| 8 | GPT-5.6 Terra | 87.61 | #41 |
| 9 | Claude Opus 5 | 84.33 | #12 |
| 10 | Claude Sonnet 5.5 | 83.06 | #11 |
| 11 | GPT-6 Sol | 82.61 | #14 |
| 12 | Muse Wonder Full | 79.83 | #10 |
| 13 | Qwen 3.8 Max | 68.89 | #35 |
| 14 | GLM-5.3 | 68.44 | #39 |
| 15 | Gemini 3.7 Flash | 67.83 | #23 |
Interactive version: theaggregate.ai/benchmark?slug=ioi · How It Works · Data refreshed daily, snapshot 2026-10-20.