VPCT
Each model’s result as Epoch AI’s Benchmarking Hub publishes it, run by Epoch or collected from the source it cites. Facthory has not run them.
Data: Epoch AI, ‘Capabilities & benchmarking’, epoch.ai/benchmarks, licensed CC BY 4.0. Read on 16 Nov 2026.
The chart shows the 25 best of 38 results; the table lists them all.
| # | Model | Organization | Result | Evidence |
|---|---|---|---|---|
| 1 | Gemini 3 Pro Preview API access | Google DeepMind | 91% | Log or source |
| 2 | GPT-5.2 (xhigh) API access | OpenAI | 84% | Log or source |
| 3 | Gemini 3 Flash Preview API access | Google DeepMind | 72.6% | Log or source |
| 4 | GPT-5.2 (high) API access | OpenAI | 67% | Log or source |
| 5 | GPT-5 (high) API access | OpenAI | 66% | Log or source |
| 6 | GPT-5 (medium) API access | OpenAI | 63.2% | Log or source |
| 7 | GPT-5.1 (high) API access | OpenAI | 58.7% | Log or source |
| 8 | o4-mini (medium) API access | OpenAI | 57.5% | Log or source |
| 9 | GPT-5.1 (medium) API access | OpenAI | 53.3% | Log or source |
| 10 | o3 (medium) API access | OpenAI | 52% | Log or source |
| 11 | Gemini 2.5 Pro Preview (Mar 2025) API access | Google DeepMind | 48% | Log or source |
| 12 | Gemini 2.5 Pro Preview (Jun 2025) API access | Google DeepMind | 46.4% | Log or source |
| 13 | gemini-2.5-flash-preview-09-2025 API access | Google DeepMind | 46.2% | Log or source |
| 14 | GPT-4.5 Preview (Feb 2025) API access | OpenAI | 45% | Log or source |
| 15 | gemini-robotics-er-1.5-preview | – | 40.8% | Log or source |
| 16 | Gemini 2.5 Pro Preview (Jun 2025) API access | Google DeepMind | 40.5% | Log or source |
| 17 | GPT-5 mini (medium) API access | OpenAI | 40.2% | Log or source |
| 18 | Claude Opus 4.5 (32k thinking) API access | Anthropic | 40% | Log or source |
| 19 | GPT-4o (Nov 2024) API access | OpenAI | 40% | Log or source |
| 20 | Claude Sonnet 4.5 (32k thinking) API access | Anthropic | 39.8% | Log or source |
| 21 | claude-3-7-sonnet-20250219 API access | Anthropic | 39% | Log or source |
| 22 | GPT-5 mini (high) API access | OpenAI | 39% | Log or source |
| 23 | Claude Opus 4 API access | Anthropic | 38% | Log or source |
| 24 | Claude Sonnet 4.5 (no thinking) API access | Anthropic | 38% | Log or source |
| 25 | Gemini 2.5 Flash Preview (Apr 2025) API access | Google DeepMind | 38% | Log or source |
| 26 | GPT-5 nano (high) API access | OpenAI | 37.2% | Log or source |
| 27 | o1 (medium) API access | OpenAI | 37% | Log or source |
| 28 | GPT-5 nano (medium) API access | OpenAI | 35.4% | Log or source |
| 29 | Claude 3.7 Sonnet (64k thinking) API access | Anthropic | 35% | Log or source |
| 30 | claude-opus-4-1-20250805 API access | Anthropic | 35% | Log or source |
| 31 | claude-sonnet-4-20250514_32K API access | Anthropic | 34% | Log or source |
| 32 | gpt-4o-mini-2024-07-18 API access | OpenAI | 34% | Log or source |
| 33 | Claude 3.5 Sonnet (Jun 2024) API access | Anthropic | 33% | Log or source |
| 34 | Claude 3.5 Sonnet (Oct 2024) API access | Anthropic | 33% | Log or source |
| 35 | claude-opus-4-1-20250805_16K API access | Anthropic | 33% | Log or source |
| 36 | Claude Opus 4 API access | Anthropic | 33% | Log or source |
| 37 | claude-sonnet-4-20250514 API access | Anthropic | 30% | Log or source |
| 38 | gemini-2.5-flash-lite-preview-09-2025 API access | Google DeepMind | 30% | Log or source |
