whatmodel.sh / KAT-Coder-Pro V1
KAT-Coder-Pro V1
↗ share
+ contribute
KAT-Coder-Pro V1 is a closed-weights language model from Kuaishou (Kwaipilot), released October 2025.
Catalogue identity notes
Arena entry 'KAT-Coder-Pro-V1' (webdev). Released 2025-10: @KwaiAICoder post 2025-10-23 announcing the official release of KAT-Coder-Pro V1. Proprietary agentic-coding model served on StreamLake (models overview updated 2026-09-24 lists KAT-Coder-Pro V1). It had an in-place update 'KAT-Coder-Pro V1 (1110)' on 2025-11-10, before the Arena first listed it (2025-12-01). The Arena organisation field is blank.
$ how it compares
The LLMs ranked either side of KAT-Coder-Pro V1, out of 1132 with a score, then more from Kuaishou (Kwaipilot).
| # | model | score | where | ||
|---|---|---|---|---|---|
| 245 |  Devstral 2 (2512) | 62±2 | local | ||
| 246 |  Gemini 2.5 Flash Preview (05-20) | 62±1 | comparison only | ||
| 247 | KAT-Coder-Pro V1 | 62±3 | api price unknown | ||
| 248 |  A.X K1 (519B-A33B) | 62±1 | local | ||
| 249 |  watt-tool-8B | 62±7 | local | ||
| more from Kuaishou (Kwaipilot) | |||||
| 99 | KAT-Coder-V2.5-Dev (35B-A3B) | 70±2 | local | ||
$ what it's good at
coding▁▁▁▃▆██▇▆▄▃▁ #158 of 469 * · 97% of its score · 1 test
Beats Gemini 2.5 Pro, Qwen3-Coder-Next (80B-A3B), Grok 4.1 Fast (Reasoning) here, though they rank higher overall. Trails QwQ-32B-Preview, Ministral 3 8B Reasoning 2512, Phi-4-reasoning 14B here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| Arena WebDev (Code Arena) | 1255.3 ✎ | ▁▂▄▃▅█▄▇▆▄▁▁ beats 14/119 | 97% |
$ where to run it
Through an API
api · StreamLake · price unknown · 256k context · www.streamlake.ai ↗
$ how the score was worked out · LMArena leaderboard dataset 97%
Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works
| source | results | share of the score |
|---|---|---|
 LMArena leaderboard dataset | 1 | 97% |
| prior · pull toward the middle | 3% | |
| weighted average | = 62.1 |
$ improve this page
Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.
data updated · sources and licences · json