| # | Model | Vendor | Best score | Runs | Last seen |
|---|---|---|---|---|---|
| 1 | o1 | OpenAI | 96.6 | 1 | 2026-07-03 |
| 2 | DeepSeek-R1 | DeepSeek | 96.3 | 1 | 2026-07-03 |
| 3 | OpenAI o1-mini | OpenAI | 93.4 | 1 | 2026-07-03 |
| 4 | DeepSeek-V3 | DeepSeek | 58.7 | 1 | 2026-07-03 |
| 5 | GPT-4o | OpenAI | 23.6 | 1 | 2026-07-03 |
| 6 | Claude 3.5 Sonnet | Anthropic | 20.3 | 1 | 2026-07-03 |
Best tracked score per model (default configuration; source-attributed and verification-tiered). Open a model for its full benchmark surface, provenance and pricing.