GPT OSS 120B

Release Date: 8/5/2025

Vals Index

Accuracy (Vals Index)

36.07% ± 1.75

Latency (Vals Index)

202.33s

Cost/Test (Vals Index)

$0.08

Context Window

128k

Max Output Tokens

33k

Input Modality

Hyperparameter settings
Default Provider : Fireworks AI

Temperature

Default

Top P

Default

Top K

Default

Max Output Tokens

32,768

Show rankings only among open weight models

Benchmarks

Accuracy

Rankings

0.0%

± 1.75
41/ 45

0.0%

± 0.88
48/ 50

0.0%

± 0.97
55/ 100

0.0%

± 2.24
42/ 49

0.0%

± 0.88
48/ 107

0.0%

± 1.22
18/ 96

0.0%

± 2.25
41/ 102

0.0%

± 1.05
27/ 108

0.0%

± 0.52
74/ 119

0.0%

± 0.24
39/ 95

0.0%

± 0.40
66/ 101

0.0%

± 2.11
43/ 44

0.0%

± 4.19
49/ 56
Contact us
Or send us an email at contact@vals.ai
Proprietary Benchmarks (contact us to get access)
Academic Benchmarks

Read about our methodology.