IOI - IOI 2024
Code · 2025-08-11
The IOI 2024 split of IOI. This child benchmark separates a source-reported subtask or subtrack from the parent aggregate so scores at different grains do not share one benchmark_slug.
Top models (higher is better)
| Model | Score |
|---|---|
| Claude Fable 5 | 100.0 |
| Claude Opus 5 | 100.0 |
| GPT-5.6 Sol | 85.0 |
| GPT-5.6 Luna | 78.0 |
| GPT-5.4 | 67.2 |
| GPT-5.6 Terra | 66.2 |
| GPT-5.3-Codex | 58.2 |
| Opus 4.7 | 53.2 |
| GPT-5.2 | 43.8 |
| Gemini 3 Flash Preview | 41.0 |
| Qwen3.7-Max | 40.5 |
| Gemini 3 Pro Preview | 38.5 |
| DeepSeek-V4-Pro | 31.3 |
| Gemini 3.5 Flash-Lite | 25.0 |
| GLM-5 | 23.7 |
| Grok 4 | 23.2 |
| Grok 4.20 | 22.5 |
| Qwen3 Max (rolling alias) | 21.5 |
| Opus 4.5 | 20.7 |
| Sonnet 4.5 | 19.0 |
| Gemini 2.5 Pro | 19.0 |
| Opus 4.1 | 18.7 |
| GPT-5.1-Codex-Max | 14.5 |
| GPT-5.4 Nano | 14.2 |
| DeepSeek-V3.2 | 14.0 |
| Grok 4.3 | 13.0 |
| Qwen3 Max Thinking (2026-01-23) | 12.3 |
| Kimi K2.5 | 12.2 |
| GPT-5.1 | 11.8 |
| Qwen3 Max Preview | 11.7 |
| GPT-5 | 11.0 |
| GLM-4.7 | 10.5 |
| Sonnet 4 | 9.7 |
| Grok 4 Fast | 9.3 |
| Grok 4.1 Fast | 7.7 |
| O4 Mini | 7.7 |
| MiniMax M2.5 | 7.5 |
| GPT-5.4 Mini | 6.7 |
| Haiku 4.5 | 6.5 |
| GPT-5 Mini | 6.5 |