- Intelligence
- 11.6
- Rank 389 of 748 measured models
- Coding
- 30.4
- Rank 183 of 298 measured models
- Output speed
- 214 t/s
- Rank 37 of 406 measured models
- Price/1M
- $0.26
- Rank 193 of 664 measured models
- Context window
- 131K
- Rank 744 of 1209 measured models
- Cost per task
- $0.107
- Rank 53 of 196 measured models
gpt-oss-120b (High) scores 11.6 on the Artificial Analysis Intelligence Index, ranking 389th of 748 measured models. The closest models on this index are HyperNova 60B 2605 (High, Based on gpt-oss-120b) (11.7), MiniMax M1 80k (11.7), and Nemotron Cascade 2 30B A3B (11.7). At $0.26 per million tokens (3:1 input/output blended price), it is cheaper than MiniMax M1 80k ($0.963). With an output speed of 214 t/s, it is faster than MiniMax M1 80k (67.4 t/s).
Additional benchmarks6
Design Arena · 3d29.4%
Design Arena · Codecategories33.4%
Design Arena · Dataviz43.6%
Design Arena · Gamedev40.5%
Design Arena · Uicomponent35.7%
Design Arena · Website32.5%
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.
- HyperNova 60B 2605 (High, Based on gpt-oss-120b)
Multiverse Computing · — · —
11.7+0.1——Compare - MiniMax M1 80k
MiniMax · $0.963 · 67.4 t/s
11.7+0.1$0.96367.4 t/sCompare - Nemotron Cascade 2 30B A3B
NVIDIA · — · —
11.7+0.1——Compare - K2 Think V2
Institute of Foundation Models · — · —
11.5−0.1——Compare
