DeepInfra
gemma-3-4b-it
Made by Google DeepMind. Released 2025-03-12. Accessibility not recorded. Cheapest standard rate we hold: $0.10 per million output tokens, no context split.
What it costs
| Tier | Kind | In | Out | Cached in | From |
|---|---|---|---|---|---|
| standard | text | $0.05 | $0.10 | — | 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 |
|---|---|---|---|---|
| chess puzzles | Epoch AI | 0.0±0.00 | 2026-08-28 | — |
| gpqa diamond | Epoch AI | 0.2±0.02 | 2026-08-28 | — |
| otis mock aime 2024 2025near its ceiling | Epoch AI | 0.1±0.03 | 2026-08-28 | — |
| dtbenchnear its ceiling | collected by Epoch AI | 0.5 | — | — |
| lmca | collected by Epoch AI | 2.8 | — | — |
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 DeepInfra