REVIEW 28 cited by
Text-to-CadQuery: A New Paradigm for CAD Generation with Scalable Large Model Capabilities
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Text-to-CadQuery: A New Paradigm for CAD Generation with Scalable Large Model Capabilities
read the original abstract
Computer-aided design (CAD) is fundamental to modern engineering and manufacturing, but creating CAD models still requires expert knowledge and specialized software. Recent advances in large language models (LLMs) open up the possibility of generative CAD, where natural language is directly translated into parametric 3D models. However, most existing methods generate task-specific command sequences that pretrained models cannot directly handle. These sequences must be converted into CAD representations such as CAD vectors before a 3D model can be produced, which requires training models from scratch and adds unnecessary complexity. To tackle this issue, we propose generating CadQuery code directly from text, leveraging the strengths of pretrained LLMs to produce 3D models without intermediate representations, using this Python-based scripting language. Since LLMs already excel at Python generation and spatial reasoning, fine-tuning them on Text-to-CadQuery data proves highly effective. Given that these capabilities typically improve with scale, we hypothesize that larger models will perform better after fine-tuning. To enable this, we augment the Text2CAD dataset with 170,000 CadQuery annotations. We fine-tune six open-source LLMs of varying sizes and observe consistent improvements. Our best model achieves a top-1 exact match of 69.3%, up from 58.8%, and reduces Chamfer Distance by 48.6%. Project page: https://github.com/Text-to-CadQuery/Text-to-CadQuery.
Forward citations
Cited by 28 Pith papers
-
neuralCAD-Edit: An Expert Benchmark for Multimodal-Instructed 3D CAD Model Editing
neuralCAD-Edit benchmark shows even the best foundation model (GPT 5.2) scores 53% lower than human CAD experts in acceptance trials for multimodal-instructed 3D model edits.
-
IterCAD: An Iterative Multimodal Agent for Visually-Grounded CAD Generation and Editing
IterCAD introduces a closed-loop multimodal agent for CAD generation and editing, trained via progressive SFT and geometry-aware RL with viable-prefix masking, and evaluated on IterCAD-Bench using a new CD-TR curve an...
-
P3D-Bench: Benchmarking MLLMs for Parametric 3D Generation and Structural Reasoning
P3D-Bench is a benchmark with three task families that scores MLLMs on generating executable parametric 3D programs, finding failures in precise geometry and part assembly.
-
UniCAD: A Unified Benchmark and Universal Model for Multi-Modal Multi-Task CAD
UniCAD supplies a unified multi-modal benchmark and an end-to-end MLLM that performs reconstruction, generation, and QA on CAD data, reporting SOTA results on UniCAD and Fusion360.
-
Text2CAD-Bench: A Benchmark for LLM-based Text-to-Parametric CAD Generation
Text2CAD-Bench supplies 600 dual-prompt examples across four geometric and domain levels to test LLMs on text-to-parametric CAD, finding solid basic performance but sharp drops on complex topology and advanced features.
-
CADBench: A Multimodal Benchmark for AI-Assisted CAD Program Generation
CADBench is a new multimodal benchmark for CAD program generation that combines 18k samples from DeepCAD, Fusion 360, ABC, MCB, and Objaverse across clean/noisy meshes and various renders, used to test 11 models and r...
-
ArtiCAD: Articulated CAD Assembly Design via Multi-Agent Code Generation
ArtiCAD presents the first training-free multi-agent framework that generates articulated, editable CAD assemblies from text or images by predicting assembly relationships early and using validation with rollback.
-
PR-CAD: Progressive Refinement for Unified Controllable and Faithful Text-to-CAD Generation with Large Language Models
PR-CAD unifies text-to-CAD generation and editing via progressive refinement with LLMs, a new interaction dataset, and RL-enhanced reasoning to achieve better controllability and faithfulness.
-
HistCAD: A Constraint-Aware Parametric History-Based CAD Representation, Dataset, and Benchmark with Industrial Complexity
HistCAD provides a constraint-aware parametric CAD representation, a dataset of 170k industrial sequences, and an editability benchmark with metrics ER, cPCSR, and OES to evaluate preservation of design intent.
-
Nova3D: Code-Native Generation of Programmable 3D Assets
Nova3D generates 3D assets as executable Blender source, yielding named parts, assembly hierarchies, measurable constraints, and native joints that mesh-native generators do not expose.
-
ArtisanCAD: An Industrial-Level CAD Agent with Expert-Grounded Knowledge Distillation
CAD-IR scaffolds ambiguous text into executable CATIA procedures and, with expert-distilled skills, produces editable B-Rep variants of complex automotive parts.
-
ASSEMCAD: Production-Ready CAD Assembly Generation from Natural Language
AssemCAD generates production-oriented CAD assemblies from language via axiom-grounded specs, port-mate libraries, on-demand factories, and geometry-aware verification, beating code-only baselines on AssemBench.
-
IterCAD: An Iterative Multimodal Agent for Visually-Grounded CAD Generation and Editing
IterCAD is a multimodal agent framework using progressive SFT and geometry-aware RL for CAD tasks, with a new data pipeline, IterCAD-Bench, and CD-TR metric showing outperformance in executability and precision.
-
Self-Improving CAD Generation Agents with Finite Element Analysis as Feedback
CAD generation agents are augmented with FEA feedback plus text blueprint and 21-view image signals, raising Box-IoU on S2O and Fusion360 while showing that base models produce no strict-passing FEA artifacts.
-
CADBench: A Multimodal Benchmark for AI-Assisted CAD Program Generation
CADBench is a multimodal benchmark for CAD program generation that shows specialized mesh-to-CAD models outperform general vision-language models but degrade with complexity and modality shifts.
-
Agent-Aided Design for Dynamic CAD Models
AADvark extends agent-aided CAD design to dynamic 3D assemblies with movable parts by integrating constraint solvers and visual feedback to create a verification signal for the agent.
-
COSMO-Agent: Tool-Augmented Agent for Closed-loop Optimization,Simulation,and Modeling Orchestration
COSMO-Agent trains LLMs via tool-augmented RL and a multi-constraint reward to close the CAD-CAE loop, with experiments showing small open-source models outperforming larger ones on feasibility and stability for 25 co...
-
DreamCAD: Scaling Multi-modal CAD Generation using Differentiable Parametric Surfaces
A point-supervised model represents 3D shapes as connected Bézier patches and generates editable CAD-style surfaces from text, images, or point clouds.
-
Pointer-CAD: Unifying B-Rep and Command Sequences via Pointer-based Edges & Faces Selection
Pointer-CAD unifies B-Rep geometry with command sequences via pointer-based entity selection, allowing LLMs to perform complex CAD edits while cutting topological errors from quantization.
-
STEP-LLM: Generating CAD STEP Models from Natural Language with Large Language Models
STEP-LLM fine-tunes LLMs to emit complete STEP CAD files from natural-language captions, reporting lower median Chamfer distance than Text2CAD.
-
HistCAD: A Constraint-Aware Parametric History-Based CAD Representation, Dataset, and Benchmark with Industrial Complexity
HistCAD provides a constraint-explicit CAD sequence format, a 170,236-model dataset, and an editability benchmark measuring whether parameter edits preserve design intent.
-
CADDesigner: Conceptual CAD Model Generation with a General-Purpose Agent
CADDesigner presents an LLM agent with ECIP paradigm and visual feedback loop that generates conceptual CAD models from text and sketches while storing results in a knowledge base.
-
Wrong Design Intent Is Worse Than None: A Derangement-Control Diagnosis of Header Conditioning in CAD Program Completion
A wrong design-intent header degrades CAD completion below the no-header baseline, and a derangement-trained control shows the harm comes from the learned header-program mapping.
-
Pointer-CAD v2: Plan-Then-Construct CAD Generation with Dimension-Aware Parametric Precision
Pointer-CAD v2 decouples planning from construction in LLM-based CAD generation by using a pointer mechanism to reference continuous parameters from a design plan, paired with new hierarchical accuracy metrics.
-
Self-Improving CAD Generation Agents with Finite Element Analysis as Feedback
CAD agents using finite element analysis feedback plus new text blueprint and multi-view image signals improve geometric accuracy on S2O and Fusion360 benchmarks while addressing physical validity gaps in prior genera...
-
Tool-Augmented Agent for Closed-loop Optimization,Simulation,and Modeling Orchestration
COSMO-Agent is a tool-augmented RL agent that trains LLMs to complete closed-loop CAD-CAE optimization using a multi-constraint reward and an industry dataset of 25 component categories, improving small models over la...
-
CADDesigner: Conceptual CAD Model Generation with a General-Purpose Agent
CADDesigner is an LLM agent that generates conceptual CAD models from text and sketches via requirement analysis, the ECIP paradigm, and iterative visual feedback, outperforming baselines in experiments.
-
Bridging the Sim-to-Real Gap in Semiconductor Visual Program Synthesis via Input Binarization
Binarization of SEM inputs improves VLM-based visual program synthesis for semiconductor geometries, lifting mean Dice coefficient from 0.4393 to 0.5256 on MIIC data.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.