All benchmarksIndependent evaluation · Epoch AI
CadEval
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.
| # | Model | Organization | Result | Evidence |
|---|---|---|---|---|
| 1 | o3 (medium) API access | OpenAI | 74% | Log or source |
| 2 | Gemini 2.5 Pro Preview (Mar 2025) API access | Google DeepMind | 64% | Log or source |
| 3 | o4-mini (medium) API access | OpenAI | 62% | Log or source |
| 4 | o1 (medium) API access | OpenAI | 56% | Log or source |
| 5 | claude-3-7-sonnet-20250219 API access | Anthropic | 54% | Log or source |
| 6 | o3-mini (medium) API access | OpenAI | 54% | Log or source |
| 7 | Claude 3.5 Sonnet (Oct 2024) API access | Anthropic | 48% | Log or source |
| 8 | GPT-4.1 API access | OpenAI | 42% | Log or source |
| 9 | gemini-1.5-pro-002 API access | Google DeepMind | 34% | Log or source |
| 10 | Claude 3.5 Haiku (Oct 2024) API access | Anthropic | 32% | Log or source |
| 11 | Gemini 2.0 Flash (Feb 2025) API access | Google DeepMind,Google | 30% | Log or source |
| 12 | GPT-4o (Aug 2024) API access | OpenAI | 26% | Log or source |
| 13 | Zoo Ml-Ephant | – | 20% | Log or source |
| 14 | GPT-4.1 mini API access | OpenAI | 16% | Log or source |
| 15 | claude-3-haiku-20240307 API access | Anthropic | 12% | Log or source |
