GLM 4.6

Release Date: 9/30/2025

Benchmarked by Vals Logo

Accuracy (Average)

63.11%

Latency (Average)

916.20s

Avg. Cost (In/Out)

0.6 / 2.2

Context Window

200k

Max Output Tokens

123k

Input Modality

Hyperparameter settings
Default Provider : zAI

Temperature

0.6

Top P

1

Top K

Default

Max Output Tokens

122,880

Benchmarks

Accuracy

Rankings

CorpFin

0.0%

± 0.97

51/ 92

Finance Agent (v1.1)

0.0%

± 2.79

32/ 40

TaxEval (v2)

0.0%

± 0.81

72/ 100

AIME

0.0%

± 0.60

14/ 92

GPQA

0.0%

± 2.20

46/ 95

IOI

0.0%

± 2.49

36/ 50

LiveCodeBench

0.0%

± 1.08

29/ 101

LegalBench

0.0%

± 0.52

52/ 113

MedQA

0.0%

± 0.24

34/ 95

MMLU Pro

0.0%

± 0.37

40/ 93

Terminal-Bench 2.0

0.0%

± 4.79

29/ 46

Academic Benchmarks
Proprietary Benchmarks (contact us to get access)

Join our mailing list to receive benchmark updates

Model benchmarks are seriously lacking. With Vals AI, we report how language models perform on the industry-specific tasks where they will be used.

By subscribing, I agree to Vals' Privacy Policy.