IOI — leaderboard
Based on the International Olympiad in Informatics.
Metric: Score (self-reported). Source: benchmarklist.com. Status: saturation imminent. 54 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | DeepSeek V4 Pro | 580.1 |
| 2 | Nemotron 3 Ultra | 570 |
| 3 | GLM-5.1 | 456.5 |
| 4 | Qwen 3.5 397B A17B | 441.3 |
| 5 | Claude Fable 5 (Max) | 72.25 |
| 6 | GPT-5.4 (xHigh) | 67.83 |
| 7 | GPT-5.2 (xHigh) | 54.83 |
| 8 | Claude Opus 4.7 (Max) | 47.08 |
| 9 | Qwen 3.7 Max (Max) | 46.75 |
| 10 | GPT-5.3 Codex (xHigh) | 43.83 |
| 11 | Gemini 3 Flash (Preview) (High) | 39.08 |
| 12 | Gemini 3 Pro (High) | 38.83 |
| 13 | DeepSeek V4 Pro (Max) | 35.83 |
| 14 | Grok 4.20 | 30.17 |
| 15 | Grok 4 | 26.17 |
Interactive version: theaggregate.ai/benchmark?slug=ioi · How the rankings work · Data refreshed daily, snapshot 2026-07-22.