AIME 2026
Math · 2026-02-13
This MathArena track evaluates models on the AIME 2026 problem set. Scores report the percentage of problems answered correctly.
Top models (higher is better)
| Model | Score |
|---|---|
| Opus 4.8 | 100.0 |
| GPT-5.5 | 100.0 |
| Claude Fable 5 | 99.9 |
| GPT-5.6 Sol | 99.9 |
| GPT-5.4 | 99.2 |
| Gemini 3.1 Pro Preview | 98.3 |
| GPT-5.2 | 98.3 |
| GPT-5.6 Luna | 97.6 |
| Inkling | 97.1 |
| Opus 4.6 | 96.7 |
| DeepSeek-V4-Pro | 96.7 |
| Gemini 3.6 Flash | 96.7 |
| Gemini 3 Flash Preview | 96.7 |
| GLM-5 | 96.7 |
| Kimi K3 | 96.7 |
| Step 3.5 Flash | 96.7 |
| Opus 4.7 | 95.8 |
| DeepSeek-V4-Flash | 95.8 |
| GLM-5.1 | 95.8 |
| Kimi K2.5 | 95.8 |
| Kimi K2.6 | 95.8 |
| Inkling-Small | 95.5 |
| Inkling-Small Preview | 95.1 |
| Gemini 3.5 Flash | 95.0 |
| Step 3.7 Flash | 95.0 |
| Nemotron 3 Ultra 550B A55B | 94.2 |
| DeepSeek-V3.2 | 94.2 |
| Grok 4.1 Fast | 94.2 |
| Qwen3.5 397B A17B | 94.2 |
| MiMo-V2.5 | 93.6 |
| Qwen3.5 35B A3B | 93.3 |
| Qwen3.5-9B | 92.5 |
| Gemini 3 Pro Preview | 91.7 |
| Nemotron 3 Super 120B A12B | 91.7 |
| Qwen3.5-27B | 91.7 |
| GLM-5.2 | 90.0 |
| Qwen3.5-4B | 89.7 |
| Qwen3 30B A3B Thinking 2507 | 88.3 |
| MiniMax M2.7 | 87.7 |
| Haiku 4.5 | 85.1 |