KwaiKAT
KAT-Coder-Pro V1 is KwaiKAT's most advanced agentic coding model in the KwaiKAT series. Designed specifically for agentic coding tasks, it excels in real-world software engineering scenarios, achieving a remarkable 73.4% solve rate on the SWE-Bench Verified benchmark. KAT-Coder-Pro V1 delivers top-tier coding performance and has been rigorously tested by thousands of in-house engineers. The model has been optimized for tool-use capability, multi-turn interaction, instruction following, generalization and comprehensive capabilities through a multi-stage training process, including mid-training, supervised fine-tuning (SFT), reinforcement fine-tuning (RFT), and scalable agentic RL.
KwaiKAT model releasesRank #103 across 526
Rank #227 across 436
Rank #140 across 357
Percentile score by analysis domain.
* Cost is inverted: lower input, output, and blended prices rank higher.
Higher bars mean stronger relative placement.
| Metric | Domain | Value | Rank |
|---|---|---|---|
| Artificial Analysis Intelligence Index | overall | 36.0 | #103 |
| Artificial Analysis Coding Index | coding | 18.3 | #227 |
| Artificial Analysis Math Index | math | 94.7 | #10 |
| MMLU-Pro | reasoning | 81.3% | #90 |
| reasoning |
| 76.4% |
| #157 |
| Humanity's Last Exam | reasoning | 33.4% | #25 |
| LiveCodeBench | coding | 74.7% | #49 |
| SciCode | coding, reasoning | 36.6% | #181 |
| Output Speed | speed | 114.7 tok/s | #119 |
| Time to First Token | speed | 1.70s | #209 |
| Blended Price | cost | $0.525/M | #140 |
| Input Price | cost | $0.300/M | #148 |
| Output Price | cost | $1.20/M | #142 |
| Value Index | cost, overall | 68.6 | #96 |