Updated daily at 05:00 UTC
- Released
- 6 Nov 2023
- Status
- Deprecated
- Retired
- 23 Oct 2026
- Context
- 128K
- Max output
- 4.1K
- Input per 1M tokens
- $10.00
- Output per 1M tokens
- $30.00
- Open weights
- No
- License
- proprietary
- Parameters
- —
- Knowledge cutoff
- Dec 2023
- Reasoning
- No
- Modalities
- text, image → 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 | 7.6 | 100 | 3 of 4 |
| Human preference | 27.6 | Not ranked | 2 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 | 6.0% | Independent | — | 15 Jul 2026 | Epoch AI |
| DTBench (Decision Theory Benchmark) | Accuracy (capability questions) | 61.6% | Third party | — | — | https://conceptualreasoning.ai/dtbench via Epoch AI |
| LMCA (Language Model Conceptual Argumentation) | Correlation with expert ratings (Pearson r) | 0.0978 | Third party | — | — | https://conceptualreasoning.ai/lmca via Epoch AI |
| MMLU | Accuracy | 81.3% | Third party | — | — | Stanford CRFM Leaderboard via Epoch AI |
| MMLU | Accuracy | 79.6% | Third party | — | — | Stanford CRFM Leaderboard via Epoch AI |
| SimpleBench | Score (AVG@5) | 25.1% | Third party | — | — | SimpleBench Leaderboard via Epoch AI |
Scientific reasoning
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| GPQA Diamond | Accuracy | 46.6% | Independent | — | 27 Feb 2025 | Epoch AI |
| GPQA Diamond | Accuracy | 42.4% | Independent | — | 27 Jan 2025 | Epoch AI |
| GPQA Diamond | Accuracy | 42.3% | Independent | — | 27 Jan 2025 | Epoch AI |
Mathematics
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| FrontierMath Tiers 1-3 | Accuracy | 0.7% | Independent | — | 27 Aug 2026 | Epoch AI |
| MATH Level 5 | Accuracy | 46.7% | Independent | — | 27 Feb 2025 | Epoch AI |
| MATH Level 5 | Accuracy | 40.0% | Independent | — | 27 Jan 2025 | Epoch AI |
| MATH Level 5 | Accuracy | 35.4% | Independent | — | 27 Jan 2025 | Epoch AI |
| OTIS Mock AIME 2024-2025 | Accuracy | 6.7% | Independent | — | 27 Feb 2025 | Epoch AI |
Coding & software engineering
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| METR Time Horizons | Average task success (Epoch's column) | 36.7% | Third party | — | — | METR - Measuring AI Ability to Complete Long Tasks via Epoch AI |
| METR Time Horizons | Average task success (Epoch's column) | 35.2% | Third party | — | — | METR - Measuring AI Ability to Complete Long Tasks via Epoch AI |
| METR Time Horizons | Average task success (Epoch's column) | 28.9% | Third party | — | — | https://metr.org/blog/2025-03-19-measuring-ai-ability-to-complete-long-tasks/ via Epoch AI |
| METR Time Horizons | Average task success (Epoch's column) | 27.2% | Source not stated | — | — | Epoch AI via Epoch AI |
| WeirdML (v2) | Average accuracy | 18.0% | Third party | — | — | https://htihle.github.io/weirdml.html via Epoch AI |
Human preference
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| Text Arena (arena.ai) | Arena score (Elo scale) | 1324 | Independent | — | 25 Sep 2026 | Arena |
| Text Arena (arena.ai) | Arena score (Elo scale) | 1313 | Independent | — | 25 Sep 2026 | Arena |
| Text Arena (arena.ai) | Arena score (Elo scale) | 1313 | Independent | — | 25 Sep 2026 | Arena |
| Text Arena (arena.ai) | Votes | 100105 | Independent | — | 25 Sep 2026 | Arena |
| Text Arena (arena.ai) | Votes | 98114 | Independent | — | 25 Sep 2026 | Arena |
| Text Arena (arena.ai) | Votes | 93439 | Independent | — | 25 Sep 2026 | Arena |
| Vision Arena (arena.ai) | Arena score (Elo scale) | 1112 | Independent | — | 28 Sep 2026 | Arena |
| Vision Arena (arena.ai) | Votes | 13391 | Independent | — | 28 Sep 2026 | Arena |