A 2025 International Olympiad in Informatics benchmark used to evaluate advanced algorithmic problem solving and code generation.
Browse the latest scores, model modes, release dates, and parameter sizes for IOI 2025.
Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology
| Rank | Model | License | |||
|---|---|---|---|---|---|
 Grok 4 Thinking Enabled | 29.20 | 2025-07-10 | Unknown | Closed | |
 GPT-5 Thinking Enabled | 29.00 | 2025-08-07 | Unknown | Closed | |
 Gemini 2.5-Pro Thinking Enabled | 15.20 | 2025-06-05 | Unknown | Closed | |
4 |  Opus 4.1 Standard Mode | 11.70 | 2025-08-06 | Unknown | Closed |
5 | Claude Sonnet 4 Thinking Enabled | 4.00 | 2025-05-23 | Unknown | Closed |
6 | Claude Sonnet 4 Standard Mode | 3.30 | 2025-05-23 | Unknown | Closed |
7 | OpenAI o4 - mini Thinking Enabled | 3.00 | 2025-04-16 | Unknown | Closed |
8 |  Kimi K2 Standard Mode | 2.00 | 2025-07-11 | 1000B | Free Commercial |
9 |  DeepSeek-V3-0324 Standard Mode | 1.70 | 2025-03-24 | 671B | Free Commercial |
