Benchmarks
Vpct
Models scored
38
Best score
0.91
Metric
Correct
Vpct is an AI evaluation tracked by Epoch AI, with 38 scored model versions on record. The scores here are reported by third parties rather than produced by Epoch AI running the evaluation itself.
Scored models span June 2024 to December 2025. The best score recorded is 0.91, measured as Correct. Models come from Google DeepMind, OpenAI, Anthropic.
Full record
- Score metric
- Correct
- Models scored
- 38
- Best score recorded
- 0.91
- Earliest model scored
- Jun 20, 2024
- Most recent model scored
- Dec 17, 2025
- Third-party reported
- Yes
Leaderboard
| 01 | gemini-3-pro-preview | Google DeepMind | 0.91 |
| 02 | gpt-5.2-2025-12-11_xhigh | OpenAI | 0.84 |
| 03 | gemini-3-flash-preview | Google DeepMind | 0.73 |
| 04 | gpt-5.2-2025-12-11_high | OpenAI | 0.67 |
| 05 | gpt-5-2025-08-07_high | OpenAI | 0.66 |
| 06 | gpt-5-2025-08-07_medium | OpenAI | 0.63 |
| 07 | gpt-5.1-2025-11-13_high | OpenAI | 0.59 |
| 08 | o4-mini-2025-04-16_medium | OpenAI | 0.57 |
| 09 | gpt-5.1-2025-11-13_medium | OpenAI | 0.53 |
| 10 | o3-2025-04-16_medium | OpenAI | 0.52 |
| 11 | gemini-2.5-pro-preview-03-25 | Google DeepMind | 0.48 |
| 12 | gemini-2.5-pro-preview-06-05 | Google DeepMind | 0.46 |
| 13 | gemini-2.5-flash-preview-09-2025 | Google DeepMind | 0.46 |
| 14 | gpt-4.5-preview-2025-02-27 | OpenAI | 0.45 |
| 15 | gemini-robotics-er-1.5-preview | 0.41 | |
| 16 | gemini-2.5-pro-preview-05-06 | Google DeepMind | 0.41 |
| 17 | gpt-5-mini-2025-08-07_medium | OpenAI | 0.4 |
| 18 | claude-opus-4-5-20251101_32K | Anthropic | 0.4 |
| 19 | gpt-4o-2024-11-20 | OpenAI | 0.4 |
| 20 | claude-sonnet-4-5-20250929_32K | Anthropic | 0.4 |
| 21 | claude-3-7-sonnet-20250219 | Anthropic | 0.39 |
| 22 | gpt-5-mini-2025-08-07_high | OpenAI | 0.39 |
| 23 | gemini-2.5-flash-preview-04-17 | Google DeepMind | 0.38 |
| 24 | claude-opus-4-20250514_16K | Anthropic | 0.38 |
| 25 | claude-sonnet-4-5-20250929 | Anthropic | 0.38 |
| 26 | gpt-5-nano-2025-08-07_high | OpenAI | 0.37 |
| 27 | o1-2024-12-17_medium | OpenAI | 0.37 |
| 28 | gpt-5-nano-2025-08-07_medium | OpenAI | 0.35 |
| 29 | claude-3-7-sonnet-20250219_64K | Anthropic | 0.35 |
| 30 | claude-opus-4-1-20250805 | Anthropic | 0.35 |
| 31 | claude-sonnet-4-20250514_32K | Anthropic | 0.34 |
| 32 | gpt-4o-mini-2024-07-18 | OpenAI | 0.34 |
| 33 | claude-opus-4-1-20250805_16K | Anthropic | 0.33 |
| 34 | claude-3-5-sonnet-20241022 | Anthropic | 0.33 |
| 35 | claude-opus-4-20250514 | Anthropic | 0.33 |
| 36 | claude-3-5-sonnet-20240620 | Anthropic | 0.33 |
| 37 | gemini-2.5-flash-lite-preview-09-2025 | Google DeepMind | 0.3 |
| 38 | claude-sonnet-4-20250514 | Anthropic | 0.3 |
More benchmarks
SourceEpoch AI, 'AI Benchmarking Hub'. Published online at epoch.ai. Retrieved 2026-07-29 from https://epoch.ai/benchmarks. Licensed under CC BY 4.0.
← All benchmarks