OpenAI
gpt-5-2025-08-07
Made by OpenAI. Released 2025-08-07. Accessibility not recorded. We hold no price for this model: no seller we can read lists it any more, so what follows is what was measured, not what it cost.
What it scores, by house
| Benchmark | House | Score | Evaluated | Scaffold |
|---|---|---|---|---|
| chess puzzles | Epoch AI | 0.2±0.04 | 2026-07-15 | low |
| chess puzzles | Epoch AI | 0.3±0.05 | 2026-07-15 | medium |
| chess puzzles | Epoch AI | 0.2±0.04 | 2026-07-15 | minimal |
| chess puzzles | Epoch AI | 0.4±0.05 | 2025-12-08 | high |
| ebr bench | Epoch AI | 0.1 | 2026-06-29 | high |
| frontiermath | Epoch AI | 0.3±0.03 | 2025-11-13 | high |
| frontiermath | Epoch AI | 0.3±0.03 | 2025-11-13 | medium |
| frontiermath tier 4 | Epoch AI | 0.1±0.05 | 2025-10-30 | high |
| frontiermath tier 4 | Epoch AI | 0.1±0.04 | 2025-08-07 | medium |
| frontiermath tier 4 v2 | Epoch AI | 0.2±0.07 | 2026-06-11 | high |
| frontiermath tiers 1 3 v2 | Epoch AI | 0.4±0.03 | 2026-08-27 | low |
| frontiermath tiers 1 3 v2 | Epoch AI | 0.2±0.02 | 2026-08-27 | minimal |
| frontiermath tiers 1 3 v2 | Epoch AI | 0.6±0.03 | 2026-06-10 | high |
| gpqa diamond | Epoch AI | 0.7±0.03 | 2026-07-20 | minimal |
| gpqa diamond | Epoch AI | 0.9±0.02 | 2025-10-29 | high |
| gpqa diamond | Epoch AI | 0.9±0.02 | 2025-08-07 | medium |
| math level 5near its ceiling | Epoch AI | 1.0±0.00 | 2025-10-29 | high |
| math level 5near its ceiling | Epoch AI | 1.0±0.00 | 2025-08-20 | medium |
| mystery game puzzles | Epoch AI | 0.1±0.03 | 2026-08-27 | minimal |
| mystery game puzzles | Epoch AI | 0.2±0.04 | 2026-08-27 | low |
| mystery game puzzles | Epoch AI | 0.1±0.04 | 2026-08-27 | medium |
| mystery game puzzles | Epoch AI | 0.2±0.04 | 2026-08-05 | high |
| otis mock aime 2024 2025near its ceiling | Epoch AI | 0.5±0.08 | 2026-07-20 | minimal |
| otis mock aime 2024 2025near its ceiling | Epoch AI | 0.9±0.04 | 2025-10-29 | high |
| otis mock aime 2024 2025near its ceiling | Epoch AI | 0.9±0.04 | 2025-08-07 | medium |
| simpleqa verified | Epoch AI | 0.5±0.02 | 2026-08-27 | high |
| swe bench verified | Epoch AI | 0.7±0.02 | 2026-02-06 | high |
| swe bench verified | Epoch AI | 0.7±0.02 | 2026-02-05 | medium |
| aider polyglot | collected by Epoch AI | 88.0 | — | high |
| aider polyglot | collected by Epoch AI | 86.7 | — | medium |
| aider polyglot | collected by Epoch AI | 81.3 | — | low |
| ale bench | collected by Epoch AI | 808 | — | minimal |
| ale bench | collected by Epoch AI | 1162 | — | high |
| algotune | collected by Epoch AI | 1.7 | — | high |
| apex agents | collected by Epoch AI | 0.3 | — | high |
| apex agents | collected by Epoch AI | 0.3 | — | unknown |
| arc aginear its ceiling | collected by Epoch AI | 0.6 | — | medium |
| arc aginear its ceiling | collected by Epoch AI | 0.1 | — | minimal |
| arc aginear its ceiling | collected by Epoch AI | 0.7 | — | high |
| arc aginear its ceiling | collected by Epoch AI | 0.4 | — | low |
| arc agi 2 | collected by Epoch AI | 0.1 | — | high |
| arc agi 2 | collected by Epoch AI | 0.0 | — | minimal |
| arc agi 2 | collected by Epoch AI | 0.0 | — | low |
| arc agi 2 | collected by Epoch AI | 0.1 | — | medium |
| balrog | collected by Epoch AI | 0.3 | — | minimal |
| critpt | collected by Epoch AI | 0.0 | — | minimal |
| critpt | collected by Epoch AI | 0.1 | — | high |
| deepresearchbench | collected by Epoch AI | 0.5 | — | low |
| deepresearchbench | collected by Epoch AI | 0.5 | — | high |
| deepresearchbench | collected by Epoch AI | 0.5 | — | medium |
| deepresearchbench | collected by Epoch AI | 0.5 | — | minimal |
| dtbenchnear its ceiling | collected by Epoch AI | 0.9 | — | high |
| enigma eval | collected by Epoch AI | 0.1 | — | unknown |
| fictionlivebench | collected by Epoch AI | 1.0 | — | medium |
| forecastbench | collected by Epoch AI | 61.4 | — | unknown |
| gdpval | collected by Epoch AI | 0.3 | — | medium |
| geobench | collected by Epoch AI | 3937 | — | medium |
| gso | collected by Epoch AI | 0.1 | — | OpenHands |
| hle | collected by Epoch AI | 0.3 | — | high |
| lech mazur writing | collected by Epoch AI | 8.6 | — | medium |
| lmca | collected by Epoch AI | 40.0 | — | high |
| metr time horizons | collected by Epoch AI | 203 | — | high |
| metr time horizons | collected by Epoch AI | 137 | — | medium |
| proofbench | collected by Epoch AI | 0.2 | — | high |
| rli | collected by Epoch AI | 0.0 | — | unknown |
| scicode | collected by Epoch AI | 0.4 | — | unknown |
| simplebench | collected by Epoch AI | 0.6 | — | high |
| terminalbench | collected by Epoch AI | 0.3 | — | Mini-SWE-Agent |
| terminalbench | collected by Epoch AI | 0.5 | — | Codex CLI |
| terminalbench | collected by Epoch AI | 0.4 | — | OpenHands |
| terminalbench | collected by Epoch AI | 0.4 | — | Terminus 2 |
| vpct | collected by Epoch AI | 0.6 | — | medium |
| vpct | collected by Epoch AI | 0.7 | — | high |
| webdev arena | collected by Epoch AI | 1419 | — | medium |
| weirdml | collected by Epoch AI | 0.6 | — | high |
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.
What retires, and when
- deprecation — 2026-12-11, replaced by gpt-5.6-sol