readable AI benchmarks (simplified)

readable AI benchmarks

Kimi

Kimi K2.6

reasoning · open weights · Apr 20, 2026

Kimi K2.6 has a Quality Score of 51.9, ranking 30th among 53 scored models. Per 1M tokens, pricing is $0.950 input, $4.00 output, and $0.160 cached input. Its Reliability Score is 53.2, ranking 21st among 53 scored models, above average (total average: 50.1). Its Value Score is 38.5, around average compared with the total Value average of 39.6.

Quality Score51.9coverage 91.2%: missing Epoch General ECI, CursorBench 3.2 and DeepSWE v1.1
Value Score38.5
Reliability53.2
Cache Discount83%

Model specification

Reasoning
reasoning
Input modalities
none
Output modalities
none
Context window
not available
Weights
open weights

Token prices USD per 1M tokens

Input
$0.950
Output
$4.00
Cached Input
$0.160
Cache write
not available
Cost per task
$0.335

Capability

Intelligence
44.2
Coding
61.8
Agentic
30.3
Omniscience
6.4
Correct
32.8%
Blended price
$0.702

Answer outcomes

Attempt rate: 59.2%

Correct 32.8%Incorrect 26.4%Abstained 40.8%

Artificial Analysis benchmarks

GDPval-AA v2
34%
τ³-Banking
21%
Terminal-Bench v2.1
66%
SciCode
53%
Humanity’s Last Exam
36%
GPQA Diamond
91%
CritPt
8%
AA-Omniscience
53%
AA-LCR
70%

Similar models