AI BENCHY
Advertise here
#53

KAT-Coder-Pro V2.5

Kwaipilot Release: 2026-07-14 Tested on: 2026-07-14 11:05 kwaipilot/kat-coder-pro-v2.5::high
~700B total (~72B active)MoEClosedEstimated
(high) (medium) (low) (none)

Summary

KAT-Coder-Pro V2.5 scores 7.7 on AI BENCHY and ranks #53. It has 10.0 reliability, a 63.5% pass rate, $0.246 total cost, and 12.30s average response time.

What makes KAT-Coder-Pro V2.5 unique: It stands out most in Combined, where it ranks #1, while Anti-AI Tricks is its weakest area at #18.

Model facts

Researched on 2026-08-12

Estimated
Parameters
~700B total (~72B active)
Architecture
MoE
Availability
Closed
License
-

Best estimate from public evidence; the vendor did not disclose every value. The active count has stronger secondary evidence than the inferred total.

Parameter source Availability source

Score

7.7

Consistency

8.1

Reliability

10.0

Total Cost (Current Price)

$0.246

Total Output Tokens

72,103

Total Input Tokens

42,750

Input Price

$0.740 / 1M

Output Price

$2.960 / 1M

Tests Correct

Wrong Tests: 10

Attempt pass rate: 63.5%

Flaky tests

5

Flaky tests had mixed outcomes across runs (at least one pass and one fail).

Response Time (avg)

12.30s

Response Time (max): 95.65s

Response Time (total): 258.34s

Wrong answer: 10

Hamster playing table tennis

Prompt: Create a detailed SVG illustration of a hamster playing table tennis.

#53 KAT-Coder-Pro V2.5

high
Cost
$0.009
Time
26.4s
Tokens
3,117 tok

Run history

Tested on Score Reliability Tests Correct Total Cost Compare
2026-08-14 03:25 Re-test 7.2 10.0 $0.506 Compare
2026-07-16 23:38 New test added 7.2 10.0 $0.482 Compare
2026-07-14 11:05 Initial run 7.7 10.0 $0.246 Current run

Price History

Historical pricing data for this model from OpenRouter.

Date Input Price Output Price
2026-07-14 11:25 $0.740 / 1M $2.960 / 1M

Charts

Choose the first model, then click a second model to open a side-by-side page.

Total Output Tokens

Score vs Total Output Tokens

Quick Compare

KAT-Coder-Pro V2.5highvsKimi K2.6mediumKAT-Coder-Pro V2.5highvsMuse Glimmer 30BhighKAT-Coder-Pro V2.5highvsClaude Opus 4.8noneKAT-Coder-Pro V2.5highvsQwen3.5 Plus 2026-04-20mediumKAT-Coder-Pro V2.5highvsDeepSeek V4 Flash 0731mediumKAT-Coder-Pro V2.5highvsQwen3.7 FlashlowKAT-Coder-Pro V2.5highvsQwen3.7 PlusnoneKAT-Coder-Pro V2.5highvsMuse Glimmer 30BmediumKAT-Coder-Pro V2.5highvsGemini 3.1 Flash LitemediumKAT-Coder-Pro V2.5highvsQwen3.5-122B-A10Bmedium

Category Breakdown

Category Score Consistency Tests Correct
Anti-AI Tricks 7.0 8.0
Coding 6.4 7.9
Combined 10.0 10.0
Data parsing and extraction 10.0 10.0
Domain specific 2.9 4.4
General Intelligence 5.1 10.0
Instructions following 9.9 10.0
Puzzle Solving 8.2 7.2
Tool Calling 10.0 10.0
Trivia 3.0 10.0

Compared models