AI BENCHY
Advertise here
#204

KAT-Coder-Air V2.5

Kwaipilot Release: 2026-07-14 Tested on: 2026-08-14 03:22 kwaipilot/kat-coder-air-v2.5::high
~35B total (~3B active)MoEClosedEstimated

Summary

KAT-Coder-Air V2.5 scores 5.6 on AI BENCHY and ranks #204. It has 9.8 reliability, a 42.4% pass rate, $0.077 total cost, and 15.87s average response time.

What makes KAT-Coder-Air V2.5 unique: It stands out most in Combined, where it ranks #1, while Data parsing and extraction is its weakest area at #11. It uses unusually many reasoning tokens, which can help explain its slower or more expensive runs.

Model facts

Researched on 2026-08-12

Estimated
Parameters
~35B total (~3B active)
Architecture
MoE
Availability
Closed
License
-

Best estimate from public evidence; the vendor did not disclose every value. Weights were announced but were not available for the exact checkpoint at research time.

Score

5.6

Consistency

8.1

Total Output Tokens

114,654

Total Input Tokens

50,423

Input Price

$0.150 / 1M

Output Price

$0.600 / 1M

Tests Correct

Wrong Tests: 15

Attempt pass rate: 42.4%

Flaky tests

5

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

Response Time (avg)

15.87s

Response Time (max): 118.35s

Response Time (total): 349.16s

Hamster playing table tennis

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

#204 KAT-Coder-Air V2.5

high
Invalid SVG
Cost
$0.000
Time
300.0s
Tokens
0 tok

Run history

Tested on Score Reliability Tests Correct Total Cost Compare
2026-08-14 03:22 Re-test 5.6 9.8 $0.077 Current run
2026-07-16 23:35 New test added 5.6 9.9 $0.077 Compare
2026-07-14 11:05 Initial run 6.2 9.7 $0.064 Compare

This run used a different benchmark suite. Keep suite changes in mind when reading historical movement.

Price History

Historical pricing data for this model from OpenRouter.

Date Input Price Output Price
2026-08-14 12:51 $0.150 / 1M $0.600 / 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

Category Breakdown

Category Score Consistency Tests Correct
Anti-AI Tricks 6.9 5.8
Coding 4.3 7.3
Combined 6.5 10.0
Data parsing and extraction 6.5 10.0
Domain specific 3.0 10.0
General Intelligence 5.1 10.0
Instructions following 9.8 10.0
Puzzle Solving 3.5 4.4
Tool Calling 10.0 10.0
Trivia 3.0 10.0

Compared models