KAT-Coder-Air V2.5: benchmark results
Provider: Kuaishou. Released 2026-07-10. Access: API.
Aggregate Rating 1622 ± 37, rank #816 of 10916 rated models, from 8 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Tinybird AI SQL Benchmark - Exactness | 52.85 | Result exactness vs human reference queries (0-100) | 76.2 |
| Tinybird AI SQL Benchmark - Success Rate | 100 | Questions answered with a valid query within 3 retries (%) | 72 |
| AI Chess Leaderboard (Continuation) | 797 | Elo | 68.1 |
| Wolfram LLM Benchmarking Project | 41.6 | Correct Functionality (%) | 44.5 |
| Tinybird AI SQL Benchmark - First-Attempt Success Rate | 92 | Questions answered with a valid query on the first attempt ( | 42.2 |
| SvelteBench | 85.6 | Average pass@1 (%) | 40 |
| AI Chess Leaderboard (Reasoning) | 616 | Elo | 35 |
| ObviousBench | 74.31 | Answer pass³ (%) | 26.7 |
Interactive version: theaggregate.ai/model?slug=kat-coder-air-v2-5 · How It Works · Data refreshed daily, snapshot 2026-11-08.