CADTestBench
Models generate mechanical parts from text. Executable tests check the geometry and topology of each solid. A part passes when it satisfies every test.
The authors compare five language model planners on 200 parts, each with an abstract prompt and a detailed prompt. They publish separate scores for each prompt type.
- Made by
- Dimitrios Mallis and colleagues
- Released
- May 2026
- Kind
- CAD models
- Level
- University
- Scored
- Checked automatically
- Answers
- Code
- Tasks
- All public
- Measures
- Capability
- Subject
- Mechanical engineering
- Language
- English
- Interaction
- Single turn
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- CAD
- Strongest model tested
- GPT-5.2 #51 on the capability index
- Last model added
- February 2026
- Scores refreshed
- November 20th, 2026
Claude Sonnet 4.6 leads with 81% among models released by February 2026 Claude Sonnet 4.6 leads with 81% among models released by February 2026
Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. This view shows only models we track, each at its best.
The newest model with a score here came out in February 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 16 more
- 1 Claude Sonnet 4.6 Abstract 81%
- 2 GPT-5.2 Abstract 72%
- 3 GPT-4.1 Abstract 63%
- 4
Qwen3-Coder-480B-A35B Abstract 60.5%
- 5
DeepSeek-V3 Mar 2025, Abstract 39%
- 6 Text2CAD Abstract 8.50%
Source: Official CADTestBench leaderboard
- 1 Claude Sonnet 4.6 Abstract 81%
- 2 GPT-5.2 Abstract 72%
- 3 GPT-4.1 Abstract 63%
- 4
Qwen3-Coder-480B-A35B Abstract 60.5%
- 5
DeepSeek-V3 Mar 2025, Abstract 39%
Source: Official CADTestBench leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Claude Sonnet 4.6 | 81% | Abstract, CADTests + Log | Feb 2026 | Official leaderboard |
| 2 | GPT-5.2 | 72% | Abstract, CADTests | Dec 2025 | Official leaderboard |
| 3 | GPT-4.1 | 63% | Abstract, CadCodeVerify | Apr 2025 | Official leaderboard |
| 4 | | 60.5% | Abstract, ReAct | Jul 2025 | Official leaderboard |
| 5 | | 39% | Abstract, ReAct | Mar 2025 | Official leaderboard |
| 6 | Text2CAD | 8.50% | Abstract | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Claude Sonnet 4.6 | 81% | Abstract, CADTests + Log | Feb 2026 | Official leaderboard |
| 2 | GPT-5.2 | 72% | Abstract, CADTests | Dec 2025 | Official leaderboard |
| 3 | GPT-4.1 | 63% | Abstract, CadCodeVerify | Apr 2025 | Official leaderboard |
| 4 | | 60.5% | Abstract, ReAct | Jul 2025 | Official leaderboard |
| 5 | | 39% | Abstract, ReAct | Mar 2025 | Official leaderboard |
Scores at each effort level
Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.
Source: Official CADTestBench leaderboard
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| Claude Sonnet 4.6 | Abstract | CADTests + Log | 81% | ||
| Claude Sonnet 4.6 | Abstract | CADTests | 76.5% | ||
| GPT-5.2 | Abstract | CADTests | 72% | ||
| Claude Sonnet 4.6 | Abstract | ReAct | 71.5% | ||
| GPT-5.2 | Abstract | CADTests + Log | 70% | ||
| GPT-5.2 | Abstract | ReAct | 69.5% | ||
| GPT-5.2 | Abstract | ReAct + Image | 69.5% | ||
| Claude Sonnet 4.6 | Abstract | 10-shots | 64% | ||
| Claude Sonnet 4.6 | Abstract | Skilled | 64% | ||
| GPT-4.1 | Abstract | CadCodeVerify | 63% | ||
| Claude Sonnet 4.6 | Detailed | CADTests + Log | 62.5% | ||
| Abstract | ReAct | 60.5% | |||
| GPT-5.2 | Abstract | Skilled | 60% | ||
| Claude Sonnet 4.6 | Detailed | CADTests | 59% | ||
| GPT-5.2 | Abstract | 10-shots | 58% | ||
| Claude Sonnet 4.6 | Detailed | ReAct | 58% | ||
| GPT-5.2 | Detailed | CADTests + Log | 51.5% | ||
| Claude Sonnet 4.6 | Detailed | 10-shots | 51% | ||
| Claude Sonnet 4.6 | Detailed | Skilled | 50.5% | ||
| GPT-5.2 | Detailed | CADTests | 50% | ||
| GPT-5.2 | Detailed | ReAct + Image | 48.5% | ||
| GPT-5.2 | Detailed | ReAct | 48% | ||
| GPT-5.2 | Detailed | 10-shots | 46.5% | ||
| GPT-5.2 | Detailed | Skilled | 43% | ||
| Detailed | ReAct | 42.5% | |||
| GPT-4.1 | Detailed | CadCodeVerify | 41% | ||
| Abstract | ReAct | 39% | |||
| Detailed | ReAct | 37% |
Scores at each effort level
Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.
Source: Official CADTestBench leaderboard
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| Claude Sonnet 4.6 | Abstract | CADTests + Log | 81% | ||
| Claude Sonnet 4.6 | Abstract | CADTests | 76.5% | ||
| GPT-5.2 | Abstract | CADTests | 72% | ||
| Claude Sonnet 4.6 | Abstract | ReAct | 71.5% | ||
| GPT-5.2 | Abstract | CADTests + Log | 70% | ||
| GPT-5.2 | Abstract | ReAct | 69.5% | ||
| GPT-5.2 | Abstract | ReAct + Image | 69.5% | ||
| Claude Sonnet 4.6 | Abstract | 10-shots | 64% | ||
| Claude Sonnet 4.6 | Abstract | Skilled | 64% | ||
| GPT-4.1 | Abstract | CadCodeVerify | 63% | ||
| Claude Sonnet 4.6 | Detailed | CADTests + Log | 62.5% | ||
| Abstract | ReAct | 60.5% | |||
| GPT-5.2 | Abstract | Skilled | 60% | ||
| Claude Sonnet 4.6 | Detailed | CADTests | 59% | ||
| GPT-5.2 | Abstract | 10-shots | 58% | ||
| Claude Sonnet 4.6 | Detailed | ReAct | 58% | ||
| GPT-5.2 | Detailed | CADTests + Log | 51.5% | ||
| Claude Sonnet 4.6 | Detailed | 10-shots | 51% | ||
| Claude Sonnet 4.6 | Detailed | Skilled | 50.5% | ||
| GPT-5.2 | Detailed | CADTests | 50% | ||
| GPT-5.2 | Detailed | ReAct + Image | 48.5% | ||
| GPT-5.2 | Detailed | ReAct | 48% | ||
| GPT-5.2 | Detailed | 10-shots | 46.5% | ||
| GPT-5.2 | Detailed | Skilled | 43% | ||
| Detailed | ReAct | 42.5% | |||
| GPT-4.1 | Detailed | CadCodeVerify | 41% | ||
| Abstract | ReAct | 39% | |||
| Detailed | ReAct | 37% |
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official CADTestBench leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official CADTestBench leaderboard
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
More in Engineering and CAD
- EEBench In use 13 electrical-engineering circuit and PCB design tasks graded by deterministic build checks, simulation and BOM constraints. No LLM judge.
Gemini 4 Carbon 83.9%
- 3DCodeBench In use Agentic procedural 3D modeling via code: author engine-ready 3D assets through software APIs. Graded for executability and shape fidelity.
LongCat-2.5 55.4%
- 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%
- PartBench Only in system cards Parametric mechanical part design and CAD-style generation, graded for geometric and manufacturing fidelity. Grok 4.6 58.5%
- Razavi-Bench Not in model cards Models explain what happens in analog circuits and why. A language model awards up to four points per answer against a reference and a rubric.
Gemini 4 Carbon 98.3%
- CircuitRubric Not in model cards Models write analog circuit netlists from design requests. A program compares transistor connections and relative device sizes with accepted reference circuits. Claude Opus 4.8 96%
Data audit
9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 9 months ago, in February 2026, for Claude Sonnet 4.6. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
- Pass Independent runs Scores from the official leaderboard (6 entries).
- Pass Models scored 5 models we track, plus 1 other entry on its leaderboard.
- Warning Newest score The newest came 9 months ago, in February 2026, for Claude Sonnet 4.6.
- Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5.2 (#51).
- Pass Headroom The best is 81%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on dimitrismallis.github.io: 28 rows for models we track and 1 other entry at the last read.
- Pass Last read November 20th, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.