CadEval
o3 74% leads since Apr 2025 Not in model cards
Models write OpenSCAD code for single mechanical parts from text descriptions. Rendered parts are checked against reference geometry, scored by pass rate.
Made by Will Patrick Released Apr 2025 CAD models All tasks public
Last model added Apr 2025
Scores refreshed 31 Oct 2026
- Level
- University Undergraduates and entry-level professionals
- Scored
- Checked automatically A program checks the result against an answer key, unit tests or the expected final state.
- Answers
- Code A program or patch run against tests.
15
Models scored in Epoch AI data
o3 leads with 74% among models released by Apr 2025
Best score per model in Epoch AI’s data. Where a model ran with several settings or harnesses, we keep its best.
The newest model with a score here came out in Apr 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6 Astra
- Claude Opus 5.5
- GPT-6.1 Terra
- GPT-6.1 Sol
- and 16 more
- 1 o3 medium 74%
- 2
Gemini 2.5 Pro Mar 2025 64% - 3 o4-mini medium 62%
- 4 o1 medium 56%
- 5 o3-mini medium 54%
- 6 Claude 3.7 Sonnet 54%
- 7 Claude 3.5 Sonnet (October 2024) 48%
- 8 GPT-4.1 42%
- 9 Gemini 1.5 Pro Sep 2024 34%
- 10 Claude 3.5 Haiku Oct 2024 32%
- 11 Gemini 2.0 Flash Feb 2025 30%
- 12 GPT-4o Aug 2024 26%
- 13 ml-elephant 20%
- 14 GPT-4.1 mini 16%
- 15 Claude 3 Haiku 12%
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | o3 | 74% | medium | Apr 2025 | Epoch AI |
| 2 | Gemini 2.5 Pro Mar 2025 | 64% | Mar 2025 | Epoch AI | |
| 3 | o4-mini | 62% | medium | Apr 2025 | Epoch AI |
| 4 | o1 | 56% | medium | Dec 2024 | Epoch AI |
| 5 | o3-mini | 54% | medium | Jan 2025 | Epoch AI |
| 6 | Claude 3.7 Sonnet | 54% | Feb 2025 | Epoch AI | |
| 7 | Claude 3.5 Sonnet (October 2024) | 48% | Oct 2024 | Epoch AI | |
| 8 | GPT-4.1 | 42% | Apr 2025 | Epoch AI | |
| 9 | Gemini 1.5 Pro Sep 2024 | 34% | Sep 2024 | Epoch AI | |
| 10 | Claude 3.5 Haiku | 32% | Oct 2024 | Oct 2024 | Epoch AI |
| 11 | Gemini 2.0 Flash Feb 2025 | 30% | Feb 2025 | Epoch AI | |
| 12 | GPT-4o Aug 2024 | 26% | Aug 2024 | Epoch AI | |
| 13 | ml-elephant | 20% | Epoch AI | ||
| 14 | GPT-4.1 mini | 16% | Apr 2025 | Epoch AI | |
| 15 | Claude 3 Haiku | 12% | Mar 2024 | Epoch AI |
2 models have held number one since its release in Apr 2025
Lab claims count from the day the card came out, independent runs from the model’s release.
Google DeepMind OpenAI
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Run or collected by Epoch AI
Best so far
No lab model card reports it yet
We read the headline results of 479 launch posts from 51 labs, and none of them list this benchmark.
More in Engineering and CAD
- BenchCAD In use Programmatic CAD: 17,900 execution-verified CadQuery programs over 106 industrial part families. Four matched tasks scored on geometry. GPT-6 Astra 95.9%
- EEBench In use 13 electrical-engineering circuit and PCB design tasks graded by deterministic build checks, simulation and BOM constraints. No LLM judge. Claude Opus 5.5 75%
- Parametric CAD Bench In use Agents author editable parametric FreeCAD part designs, graded against held-back references for geometry and spec consistency. Claude Opus 5.5 61%
- CADGenBench Only in system cards Agents construct CAD models from design prompts. The generation split is graded for executability and geometric correctness. Claude Fable 5 45.1%
- 3DCodeBench Only in system cards Agentic procedural 3D modeling via code: author engine-ready 3D assets through software APIs. Graded for executability and shape fidelity. Grok 4.6 54%
- PartBench Only in system cards Parametric mechanical part design and CAD-style generation, graded for geometric and manufacturing fidelity. Grok 4.6 58.5%