Cloudflare
gpt-oss-120b
Made by OpenAI. Released 2025-08-05. Accessibility not recorded. Cheapest standard rate we hold: $0.75 per million output tokens, no context split.
What it costs
| Tier | Kind | In | Out | Cached in | From |
|---|---|---|---|---|---|
| standard | text | $0.35 | $0.75 | — | registry |
per million tokens, US dollars. * marks a rate from a single registry or two that disagree — the best available number rather than a confirmed one. Every other row is confirmed against a seller or a licensed index.
What it scores, by house
| Benchmark | House | Score | Evaluated | Scaffold |
|---|---|---|---|---|
| mystery game puzzles | Epoch AI | 0.0±0.00 | 2026-08-27 | high |
| mystery game puzzles | Epoch AI | 0.0±0.01 | 2026-08-27 | medium |
| aider polyglot | collected by Epoch AI | 41.8 | — | high |
| ale bench | collected by Epoch AI | 576 | — | — |
| apex agents | collected by Epoch AI | 0.1 | — | — |
| dtbenchnear its ceiling | collected by Epoch AI | 0.8 | — | high |
| lech mazur writing | collected by Epoch AI | 7.7 | — | — |
| lmca | collected by Epoch AI | 22.1 | — | high |
| metr time horizons | collected by Epoch AI | 42.0 | — | — |
| simplebench | collected by Epoch AI | 0.2 | — | — |
| surface evolver bench | collected by Epoch AI | 0.3 | — | — |
| terminalbench | collected by Epoch AI | 0.2 | — | Terminus 2 |
| terminalbench | collected by Epoch AI | 0.1 | — | Mini-SWE-Agent |
| vending bench 2 | collected by Epoch AI | -21.5 | — | — |
| weirdml | collected by Epoch AI | 0.5 | — | high |
| weirdml | collected by Epoch AI | 0.4 | — | medium |
These scores are not comparable between houses and are not added up. A score is dated by when it was evaluated, not when the model was released, and a score without its scaffold is not a measurement — which is why both are printed.
Every rate we hold · Why there is no single best model · Everything from Cloudflare