Updated daily at 05:00 UTC
Vision-language
o3-mini
- Released
- 31 Jan 2025
- Status
- Deprecated
- Retired
- 23 Oct 2026
- Context
- 200K
- Max output
- 100K
- Input per 1M tokens
- $1.10
- Output per 1M tokens
- $4.40
- Open weights
- No
- License
- proprietary
- Parameters
- —
- Knowledge cutoff
- Oct 2023
- Reasoning
- Yes
- Modalities
- text → text
Category scores
Out of 100, from independent and third-party results only, by method category-2026-09-29.1.
| Category | Score | Rank | Coverage |
|---|---|---|---|
| Knowledge & reasoning | 14.1 | 88 | 3 of 4 |
| Coding & software engineering | 30.9 | Not ranked | 2 of 5 |
| Factuality & hallucination | 15.3 | 62 | 1 of 1 |
| Human preference | 44.8 | Not ranked | 1 of 5 |
Benchmark results
Every result we hold for this model, with who ran it and where it was published.
Knowledge & reasoning
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| Chess Puzzles | Puzzles solved | 17.0% | Independent | high | 8 Dec 2025 | Epoch AI |
| Chess Puzzles | Puzzles solved | 9.0% | Independent | medium | 7 Aug 2026 | Epoch AI |
| Chess Puzzles | Puzzles solved | 6.0% | Independent | low | 15 Jul 2026 | Epoch AI |
| DTBench (Decision Theory Benchmark) | Accuracy (capability questions) | 68.8% | Third party | high | — | https://conceptualreasoning.ai/dtbench via Epoch AI |
| LMCA (Language Model Conceptual Argumentation) | Correlation with expert ratings (Pearson r) | 0.1897 | Third party | high | — | https://conceptualreasoning.ai/lmca via Epoch AI |
| Mystery Game Puzzles | Puzzles solved | 7.0% | Independent | high | 27 Aug 2026 | Epoch AI |
| SimpleBench | Score (AVG@5) | 22.8% | Third party | high | — | SimpleBench Leaderboard via Epoch AI |
Scientific reasoning
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| GPQA Diamond | Accuracy | 77.0% | Independent | high | 13 Feb 2025 | Epoch AI |
| GPQA Diamond | Accuracy | 74.3% | Independent | medium | 31 Jan 2025 | Epoch AI |
| GPQA Diamond | Accuracy | 68.2% | Independent | low | 20 Jul 2026 | Epoch AI |
Mathematics
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| FrontierMath Tier 4 | Accuracy | 0.0% | Independent | high | 11 Jun 2026 | Epoch AI |
| FrontierMath Tier 4 (v1) | Accuracy | 4.2% | Independent | high | 1 Jul 2025 | Epoch AI |
| FrontierMath Tiers 1-3 | Accuracy | 18.6% | Independent | high | 11 Jun 2026 | Epoch AI |
| FrontierMath Tiers 1-3 | Accuracy | 10.5% | Independent | medium | 27 Aug 2026 | Epoch AI |
| FrontierMath Tiers 1-3 | Accuracy | 3.9% | Independent | low | 27 Aug 2026 | Epoch AI |
| FrontierMath Tiers 1-3 (v1) | Accuracy | 12.4% | Independent | high | 16 Nov 2025 | Epoch AI |
| FrontierMath Tiers 1-3 (v1) | Accuracy | 8.1% | Independent | medium | 6 Mar 2025 | Epoch AI |
| MATH Level 5 | Accuracy | 96.5% | Independent | high | 13 Feb 2025 | Epoch AI |
| MATH Level 5 | Accuracy | 95.2% | Independent | medium | 31 Jan 2025 | Epoch AI |
| OTIS Mock AIME 2024-2025 | Accuracy | 76.9% | Independent | high | 27 Feb 2025 | Epoch AI |
| OTIS Mock AIME 2024-2025 | Accuracy | 63.9% | Independent | medium | 25 Feb 2025 | Epoch AI |
| OTIS Mock AIME 2024-2025 | Accuracy | 44.4% | Independent | low | 20 Jul 2026 | Epoch AI |
Coding & software engineering
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| Aider Polyglot | Percent correct | 60.4% | Third party | high | — | https://aider.chat/docs/leaderboards/ via Epoch AI |
| Aider Polyglot | Percent correct | 53.8% | Third party | medium | — | https://aider.chat/docs/leaderboards/ via Epoch AI |
| CadEval | Overall pass rate | 54.0% | Third party | medium | — | CadEval Dashboard via Epoch AI |
| Cybench | Unguided challenges solved | 22.5% | Third party | medium | — | Cybench leaderboard via Epoch AI |
| GSO | OPT@1 | 1.3% | Third party | low | — | GSO Leaderboard via Epoch AI |
| GSO | OPT@1 | 1.3% | Third party | high | — | GSO Leaderboard via Epoch AI |
| WeirdML (v2) | Average accuracy | 43.7% | Third party | high | — | https://htihle.github.io/weirdml.html via Epoch AI |
Long context
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| Fiction.LiveBench | % correct at 16K tokens (Epoch's column) | 50.0% | Third party | medium | — | Fiction.live leaderboard via Epoch AI |
Factuality & hallucination
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| SimpleQA Verified | Accuracy (% correct) | 15.3% | Independent | high | 31 Aug 2026 | Epoch AI |
Writing
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| LLM Creative Story-Writing Benchmark V1 (Lech Mazur) | Mean rubric score (percent of the 10-point scale) | 61.7% | Third party | high | — | lechmazur/writing Github repository via Epoch AI |
| LLM Creative Story-Writing Benchmark V1 (Lech Mazur) | Mean rubric score (percent of the 10-point scale) | 61.5% | Third party | medium | — | lechmazur/writing Github repository via Epoch AI |
Human preference
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| Text Arena (arena.ai) | Arena score (Elo scale) | 1364 | Independent | high | 25 Sep 2026 | Arena |
| Text Arena (arena.ai) | Arena score (Elo scale) | 1348 | Independent | — | 25 Sep 2026 | Arena |
| Text Arena (arena.ai) | Votes | 56572 | Independent | — | 25 Sep 2026 | Arena |
| Text Arena (arena.ai) | Votes | 18589 | Independent | high | 25 Sep 2026 | Arena |
Other o3 versions
- o3-deep-research24 Jun 2025
- o3-pro10 Jun 2025
- o316 Apr 2025