Reasoning & General on HLE
51.8ScoreKimi K2.5
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Kimi K2.5Tool Use=w/ Tools2026.02 | 51.8 | — | |
| GLM-5Tool Use=w/ Tools2026.02 | 50.4 | — | |
| GLM-4.7Tool Use=w/ Tools2026.02 | 42.8 | — | |
| DeepSeek-V3.2Tool Use=w/ Tools2026.02 | 40.8 | — | |
| Gemini 3 ProTool Use=None2026.02 | 37.2 | — | |
| GPT-5.2 (xhigh)Tool Use=None2026.02 | 35.4 | — | |
| Kimi K2.5Tool Use=None2026.02 | 31.5 | — | |
| GLM-5Tool Use=None2026.02 | 30.5 | — | |
| Claude Opus 4.5Tool Use=None2026.02 | 28.4 | — | |
| DeepSeek-V3.2Tool Use=None2026.02 | 25.1 | — | |
| GLM-4.7Tool Use=None2026.02 | 24.8 | — | |
| Claude Opus 4.5Tool Use=w/ Tools, Evaluation Set=Full set2026.02 | — | 43.4 | |
| Gemini 3 ProTool Use=w/ Tools, Evaluation Set=Full set2026.02 | — | 45.8 | |
| GPT-5.2 (xhigh)Tool Use=w/ Tools, Evaluation Set=Full set2026.02 | — | 45.5 |