Kangaroo 2025 1-2
Math · 2025-10-20
This MathArena track evaluates models on the Kangaroo 2025 1-2 problem set. Scores report the percentage of problems answered correctly.
Top models (higher is better)
| Model | Score |
|---|---|
| GPT-5.5 | 95.8 |
| GPT-5.4 | 94.8 |
| Kimi K3 | 93.8 |
| Gemini 3.5 Flash | 89.6 |
| Gemini 3 Flash Preview | 87.5 |
| Gemini 3.1 Pro Preview | 86.5 |
| Gemini 3.6 Flash | 83.3 |
| Qwen3.5 397B A17B | 81.3 |
| GPT-5.2 | 80.2 |
| Gemini 3 Pro Preview | 76.0 |
| Kimi K2.5 | 76.0 |
| Opus 4.8 | 70.8 |
| GPT-5 | 68.8 |
| GLM-4.5V | 65.6 |
| GPT-5.1 | 65.6 |
| Gemini 2.5 Pro | 64.6 |
| Sonnet 4.5 | 61.5 |
| GPT-5 Mini | 61.5 |
| Grok 4 | 61.5 |
| Grok 4.1 Fast | 60.4 |
| Opus 4.6 | 59.4 |
| Grok 4 Fast | 58.3 |
| Qwen3-VL-235B-A22B-Instruct | 58.3 |