neuralCAD-Edit
Agents edit existing CAD models from text or recorded instructions. CAD experts rate whether they would accept each edit. Software also compares the resulting geometry.
Autodesk tested three models on 192 requests. Two ASME hackathon teams published scores for the 48 text requests using a separate automatic grader.
- Made by
- Autodesk Research
- Released
- April 2026
- Kind
- CAD models
- Level
- University
- Scored
- Graded by people
- Answers
- Agent task
- Tasks
- All public
- Measures
- Capability
- Subject
- Mechanical engineering
- Language
- English
- Interaction
- Agentic
- Environment
- Desktop apps
- Items from
- Written by experts
- Refresh
- One fixed set
- Run by
- Anyone
- Occupations
- Architecture and engineering, Arts, design, entertainment, sports and media
- In views
- CAD, Agents, Vision, Audio and speech, Image, video and 3D generation
- Strongest model tested
- GPT-5.2 #51 on the capability index
- Last model added
- December 2025
- Scores refreshed
- November 20th, 2026
GPT-5.2 leads with 0.25 among models released by December 2025
Best score per model from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best.
The newest model with a score here came out in December 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 16 more
- 1 GPT-5.2 0.25
- 2
Gemini 3 Pro 0.10
- 3 Claude Sonnet 4.5 0.05
Source: Official neuralCAD-Edit leaderboard
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-5.2 | 0.25 | Dec 2025 | Official leaderboard | |
| 2 | | 0.10 | Nov 2025 | Official leaderboard | |
| 3 | Claude Sonnet 4.5 | 0.05 | Sep 2025 | Official leaderboard |
Rankings by part
We read 6 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as Chamfer-dist, Voxel-IoU and DINO-sim. GPT-5.2 leads every one of them. Arrows show how far a model moved from its overall rank among the models we track.
Chamfer-dist
3 models we track have a score here.
- 1 GPT-5.2 50.0
- 2 Claude Sonnet 4.5 ↑1 54.0
- 3
Gemini 3 Pro ↓1 110.0
Voxel-IoU
3 models we track have a score here.
- 1 GPT-5.2 0.57
- 2
Gemini 3 Pro 0.30
- 3 Claude Sonnet 4.5 0.18
DINO-sim
3 models we track have a score here.
- 1 GPT-5.2 0.66
- 2
Gemini 3 Pro 0.36
- 3 Claude Sonnet 4.5 0.25
Validity
3 models we track have a score here.
- 1 GPT-5.2 0.99
- 2
Gemini 3 Pro 0.58
- 3 Claude Sonnet 4.5 0.42
Instruction
3 models we track have a score here.
- 1 GPT-5.2 0.48
- 2
Gemini 3 Pro 0.27
- 3 Claude Sonnet 4.5 0.22
Quality
3 models we track have a score here.
- 1 GPT-5.2 0.39
- 2
Gemini 3 Pro 0.16
- 3 Claude Sonnet 4.5 0.10
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official neuralCAD-Edit 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 neuralCAD-Edit 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
7 of 8 checks pass, 1 warning. Look first at the Newest score check. The newest came 11 months ago, in December 2025, for GPT-5.2. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 1 pass 3 notes
- Pass Independent runs Scores from the official leaderboard (3 entries).
- Note Models scored Only 3 models we track.
- Warning Newest score The newest came 11 months ago, in December 2025, for GPT-5.2.
- Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5.2 (#51).
- Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on autodeskailab.github.io: 3 rows for models we track 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.