gpt-oss-20b
#129 of 274 scored.
OpenAIAug 5, 2025Open weights
Capabilities Index
Range 132.9–139.6
137.8
Context
Takes text
131K
Input
Per 1M tokens
$0.018
Output
Per 1M tokens
$0.09
Benchmarks
How it scores.
- GPQA Diamond47.7%
117th of 186
- Mock AIME65.2%
101st of 176
- Terminal-Bench3.4%
35th of 35
Everything else Epoch has for it
- DTBench46.7%
- WeirdML40.9%
- LMCA17.1%
- Chess Puzzles0.0%
On the index
Around it.
Sources