DeepInfra
DeepSeek-V4-Flash-0731
Made by DeepSeek. Released 2026-07-31. Accessibility not recorded. Cheapest standard rate we hold: $0.18 per million output tokens, no context split.
What it costs
| Tier | Kind | In | Out | Cached in | From |
|---|---|---|---|---|---|
| standard | text | $0.06 | $0.18 | $0.01 | *registry-disputed |
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 | 0.3±0.05 | 2026-08-02 | max |
| frontiermath tier 4 v2 | Epoch AI | 0.2±0.07 | 2026-08-02 | max |
| frontiermath tiers 1 3 v2 | Epoch AI | 0.6±0.03 | 2026-08-02 | max |
| gpqa diamond | Epoch AI | 0.9±0.02 | 2026-08-02 | max |
| mystery game puzzles | Epoch AI | 0.3±0.05 | 2026-08-05 | max |
| otis mock aime 2024 2025near its ceiling | Epoch AI | 0.9±0.02 | 2026-08-02 | max |
| simpleqa verified | Epoch AI | 0.3±0.01 | 2026-08-27 | max |
| ale bench | collected by Epoch AI | 1306 | — | max |
| arc aginear its ceiling | collected by Epoch AI | 0.8 | — | low |
| arc aginear its ceiling | collected by Epoch AI | 0.9 | — | max |
| arc aginear its ceiling | collected by Epoch AI | 0.1 | — | none |
| arc aginear its ceiling | collected by Epoch AI | 0.9 | — | high |
| arc agi 2 | collected by Epoch AI | 0.5 | — | low |
| arc agi 2 | collected by Epoch AI | 0.6 | — | max |
| arc agi 2 | collected by Epoch AI | 0.0 | — | none |
| arc agi 2 | collected by Epoch AI | 0.6 | — | high |
| critpt | collected by Epoch AI | 0.2 | — | max |
| dtbenchnear its ceiling | collected by Epoch AI | 0.9 | — | max |
| frontiercode | collected by Epoch AI | 0.2 | — | chisel |
| lmca | collected by Epoch AI | 41.7 | — | max |
| proofbench | collected by Epoch AI | 0.6 | — | unknown |
| scicode | collected by Epoch AI | 0.5 | — | max |
| simplebench | collected by Epoch AI | 0.6 | — | unknown |
| webdev arena | collected by Epoch AI | 1579 | — | high |
| weirdml | collected by Epoch AI | 0.6 | — | high |
| weirdml | collected by Epoch AI | 0.6 | — | max |
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