GLM 5.2

Release Date: 6/13/2026

Vals Index

Accuracy (Vals Index)

65.02% ± 1.42

Latency (Vals Index)

1122.39s

Cost/Test (Vals Index)

$2.08

Context Window

1M

Max Output Tokens

131k

Input Modality

Hyperparameter settings
Default Provider : Zhipu AI

Temperature

1

Top P

0.95

Top K

Default

Max Output Tokens

131,072

Reasoning Effort

max

Show rankings only among open weight models

Benchmarks

Accuracy

Rankings

0.0%

± 1.42
5/ 30

0.0%

± 2.00
3/ 14

0.0%

± 4.14
6/ 21

0.0%

± 0.93
13/ 116

0.0%

± 0.88
7/ 28

0.0%

± 2.17
28/ 68

0.0%

± 2.00
16/ 65

0.0%

± 4.79
9/ 42

0.0%

± 0.86
34/ 122

0.0%

± 4.79
6/ 63

0.0%

± 1.77
27/ 116

0.0%

± 1.18
72/ 122

0.0%

± 0.45
23/ 119

0.0%

± 0.33
24/ 115

0.0%

± 0.50
4/ 24

0.0%

± 1.69
3/ 60

0.0%

± 0.99
7/ 31
Contact us
Or send us an email at contact@vals.ai
Proprietary Benchmarks (contact us to get access)
Industry Partner Benchmarks
Academic Benchmarks

Read about our methodology.