Successor to K2.5 with stronger reasoning and Arena Elo; trades some agentic score for benchmark gains
| Benchmark | Score | Rank |
|---|---|---|
LiveCodeBenchvals.ai Contamination-free competitive programming (filtered by cutoff date) | 86.8% | #13 / 54 |
Arena EloArtificial Analysis Human preference ranking via blind comparisons | 1483 | #14 / 51 |
MMLU-Provals.ai Harder 10-option successor to MMLU; more reasoning-focused | 87.6% | #19 / 52 |
GPQAArtificial Analysis PhD-level science questions even experts struggle with | 91.1% | #21 / 81 |
TerminalArtificial Analysis Agentic terminal coding tasks requiring multi-step execution | 43.9% | #24 / 56 |
hleArtificial Analysis | 35.9% | #25 / 69 |
MMMUArtificial Analysis College-level multimodal reasoning across 30+ disciplines | 79.4% | #27 / 53 |