KAT Coder Pro V2: benchmark results
Kwaipilot's flagship agentic-coding KAT-Coder successor (MoE, 72B active). Provider: Kuaishou. Released 2026-03-27. Access: API.
Unified ELO 1633 ± 1, rank #172 of 1392 rated models, from 26 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA Terminal-Bench Hard | 49.24 | Accuracy (%) | 94.5 |
| AA TAU-2 Bench | 89.47 | Accuracy (%) | 83.9 |
| AA IFBench | 66.67 | Accuracy (%) | 79.5 |
| AA GPQA Diamond | 85.45 | Accuracy (%) | 78.8 |
| Wolfram LLM Benchmarking Project | 53.8 | Correct Functionality (%) | 78 |
| SnakeBench | 27 | TrueSkill Rating | 77 |
| BenchmarkList ECI | 130.63 | Capability Index (ECI) | 76.1 |
| AA Long Context Reasoning | 73 | Accuracy (%) | 73.2 |
| Artificial Analysis Intelligence Index | 25.59 | Intelligence Index | 73.1 |
| AA Humanity's Last Exam | 16.13 | Accuracy (%) | 67.7 |
| SvelteBench | 91.1 | Average pass@1 (%) | 65 |
| AA Omniscience | -22.47 | Score | 62.2 |
Interactive version: theaggregate.ai/model?slug=kat-coder-pro-v2 · How It Works · Data refreshed daily, snapshot 2026-09-05.