Kimi K2 Thinking

Release Date: 11/6/2025

Accuracy (Average)

66.56%

Latency (Average)

581.05s

Avg. Cost (In/Out)

0.6 / 2.5

Context Window

256k

Max Output Tokens

32k

Input Modality

Hyperparameter settings
Default Provider : Moonshot AI

Temperature

0.6

Top P

Default

Top K

Default

Max Output Tokens

32,000

Show rankings only among open weight models

Benchmarks

Accuracy

Rankings

0.0%

± 0.96
44/ 105

0.0%

± 0.88
48/ 111

0.0%

± 2.18
45/ 105

0.0%

± 1.16
70/ 110

0.0%

± 0.46
55/ 108

0.0%

± 0.40
54/ 104

0.0%

± 2.19
41/ 48

0.0%

± 5.15
37/ 62
Contact us
Or send us an email at contact@vals.ai
Proprietary Benchmarks (contact us to get access)
Academic Benchmarks

Read about our methodology.