CadEval
Rank #8 of 51 · 14 models scored · top score 74
unknown
Strongest on discrimination, weakest on saturation headroom. Caveat: the public test set leaves contamination on the table.
Weighted composite. Each signal is computed from the source-backed leaderboard below.
Does it still separate models?
How far from ceiling / clustered at the top?
Public vs held-out; training-leak risk.
Apples-to-apples, or mixed / vendor-optimized?
How old is the benchmark?
| # | Model | Score | Evidence | Measured |
|---|---|---|---|---|
| 1 | o3 | 74 | unverified | 2024-12-20 |
| 2 | Gemini 2.5 Pro (Mar 2025) | 64 | unverified | 2025-03-25 |
| 3 | o4-mini | 62 | unverified | 2025-04-16 |
| 4 | o1 | 56 | unverified | 2024-12-05 |
| 5 | Claude 3.7 Sonnet | 54 | unverified | 2025-02-24 |
| 6 | o3-mini | 54 | unverified | 2025-01-31 |
| 7 | Claude 3.5 Sonnet (October 2024) | 48 | unverified | 2024-10-22 |
| 8 | GPT-4.1 | 42 | unverified | 2025-04-14 |
| 9 | Gemini 1.5 Pro (Sept 2024) | 34 | unverified | 2024-09-24 |
| 10 | Claude 3.5 Haiku | 32 | unverified | 2024-10-22 |
| 11 | Gemini 2.0 Flash (Feb 2025) | 30 | unverified | 2024-12-11 |
| 12 | GPT-4o (Aug 2024) | 26 | unverified | 2024-05-13 |
| 13 | GPT-4.1 mini | 16 | unverified | 2025-04-14 |
| 14 | Claude 3 Haiku | 12 | unverified | 2024-03-04 |