Kimi K3 scores 50.6 on the Respan Index (#21), ahead of Qwen3.7-Max at 40.5 (#45). Kimi K3 leads on 12 of the 17 benchmarks both report. Qwen3.7-Max is 1.6x cheaper per token at list price. Kimi K3 has the larger context window (1.0M tokens).
* Estimated: no published scores in that category. How the Respan Index works
| 62.2% |
| 43.6% |
| LiveBench Coding (LiveBench) | 81.4% | 74.2% |
| LiveBench Data Analysis (LiveBench) | 78.7% | 71.8% |
| LiveBench Instruction Following (LiveBench) | 71.4% | 74% |
| LiveBench Language (LiveBench) | 85.5% | 79.7% |
| LiveBench Mathematics (LiveBench) | 84.4% | 85.2% |
| LiveBench Reasoning (LiveBench) | 90.7% | 83.3% |
| Mystery Game Puzzles (Epoch AI) | 26% | 32% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 97.2% | 95.6% |
| SimpleQA Verified | 50.6% | 55.8% |
| LMArena Agent (LMArena) | 0.0418 | -0.0513 |
| LMArena WebDev (LMArena) | 1658 | 1515 |
| SimpleBench (SimpleBench) | 60.7% | 70.4% |
Higher is better. Each score links to where it was published.