Competitive Programming Benchmark (TACO, LeetCode, USACO, CodeContests, IOI) 100
75Acceptance RateGemini 3.1 Pro
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Gemini 3.1 Pro2026.04 | 75 | 7 | 64.3 | |
| Claude Opus 4.62026.04 | 73 | 8 | 63.7 | |
| GPT-5.42026.04 | 72 | 7 | 63 | |
| Kimi K2.52026.04 | 65 | 5 | 53.3 | |
| DeepSeek V3.22026.04 | 65 | 4 | 52.7 | |
| Qwen 3.5-397B2026.04 | 64 | 4 | 52.3 |