Ctrl K
Back to models

KwaiKAT

KAT Coder Pro V2

  • Released
  • Closed weights
Compare
Intelligence Index

Standard setting

21.7
Estimated · Artificial Analysis v4.3.2
API price from
$0.30 in / $1.20 out / 1M tokOpenRouter · AtlasCloud · FP8
Context window
256K tok

Performance profile

ReasoningRank #39 of 107
85.5%
GPQA Diamond

Capabilities

Context window

Modalities

Measured setting

Integration

Reasoning trade-offs

Source (opens in a new tab)
Intelligence Index v4.3.2
StandardEstimated
Intelligence Index
21.7

Pricing & access

API route
OpenRouter · AtlasCloud · FP8API
Tier & contextStandard

USD per 1M tokens

Input
$0.30
Output
$1.20
Cached input
$0.06

Benchmarks

Full leaderboardSource (opens in a new tab)

Humanity’s Last Exam

Frontier reasoning

KAT Coder Pro V2 · Standard#74 of 126

  1. 1Claude Opus 5.5Max61.4%
  2. 2Claude Fable 5.1Max59.1%
  3. 3Gemini 4 ArgonHigh57.1%
  5. 71K2 Horizon 7BReasoning18.2%
  6. 72G9v3-39A5BReasoning17.5%
  7. 73HyperNova 60B 2605High16.6%
  8. 74KAT Coder Pro V2Standard16.1%
  9. 75Trinity Large ThinkingReasoning15.8%
  10. 76Gemma 4 12BReasoning15.7%
  11. 77Qwen3.5 9BReasoning14.9%

Peers use their best measured result on this benchmark.