K2-Think: benchmark results
Provider: LLM360. Released 2025-09-10. Access: Open.
Aggregate Rating 1586 ± 33, rank #1168 of 10916 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| BRIDGE Medical Leaderboard - CoT | 35.35 | Average Performance (%) | 62.2 |
| BRIDGE Medical Leaderboard - Zero-Shot | 37.81 | Average Performance (%) | 61.3 |
| ReasonScape R12 | 806.78 | ReasonScore | 46.7 |
| MathArena - AIME 2025 | 83.33 | Accuracy (%) | 45.9 |
| BRIDGE Medical Leaderboard - Few-Shot | 42.68 | Average Performance (%) | 43.7 |
| MathArena - HMMT Feb 2025 | 65 | Accuracy (%) | 34.7 |
| MathArena - SMT 2025 | 79.72 | Accuracy (%) | 31 |
| MathArena - BRUMO 2025 | 83.33 | Accuracy (%) | 23.9 |
| MathArena - CMIMC 2025 | 68.12 | Accuracy (%) | 21.4 |
| IUMB | 16.7 | Score (%) | 9.3 |
Interactive version: theaggregate.ai/model?slug=k2-think · How It Works · Data refreshed daily, snapshot 2026-11-08.