Kimi K3.1 Pro (High): benchmark results
Provider: Moonshot. Released 2026-10-23. Access: Open.
Aggregate Rating 1858 ± 20, rank #72 of 2170 rated models, from 19 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Maze-Bench | 73.09 | Mean Score | 90.7 |
| AI Coding Daily (OpenCode) - Total | 58.46 | Total points (max 70) | 90.5 |
| Creative Writing (Lechmazur) | 2.72 | Mean Score | 88.1 |
| Lech Mazur Writing | 2.72 | Comparison Score (points) | 88.1 |
| DuelLab GameBench 2 - Reasoning Variants | 71.54 | GameBench2 normalized score (0–100) | 85.1 |
| DeepsecBench | 28.71 | Recall-weighted F2 score (%) | 79.2 |
| Multi-turn Debate (Lechmazur) | 1602.6 | Bradley-Terry Rating | 76.7 |
| AI Coding Daily (OpenCode) - Laravel Code Quality | 17.7 | Laravel Code Quality (max 20) points, LLM-judged rubric scor | 76.2 |
| NonoBench | 76.67 | Standard accuracy (%), correct / non-failed attempts; versio | 58.2 |
| AI Coding Daily (OpenCode) - React-TS Code Quality | 17.33 | React-TS Code Quality (max 20) points, LLM-judged rubric sco | 57.1 |
| ASCIIBench | 1441 | Bradley–Terry score (Elo scale) | 50 |
| HLE-Diamond - Knowledge | 25.4 | Accuracy (%, 500 knowledge questions, no tools) | 41.9 |
Interactive version: theaggregate.ai/model?slug=kimi-k3-1-pro-high · How It Works · Data refreshed daily, snapshot 2026-11-05.