LLMForge is introduced, a multi-model text-to-CAD framework integrating JSON-schema validation, analytic feature scoring, mesh synthesis, and multi-round iterative refinement, studied under two critique regimes.
Abstract
Recent advances in Large Language Models (LLMs) and Vision-Language Models (VLMs) enable the automatic generation of parametric 3D designs from natural-language specifications. This chapter presents an empirical study of foundation models for automatic Computer-Aided Design (CAD) generation of mechanical parts, using a unified evaluation pipeline and a curated benchmark of 97 engineering design problems. We introduce LLMForge, a multi-model text-to-CAD framework integrating JSON-schema validation, analytic feature scoring, mesh synthesis, and multi-round iterative refinement, studied under two critique regimes. IterTracer uses a Phong-shaded ray-trace renderer with analytic visual metrics (silhouette IoU, hole visibility, edge clearance, aspect-ratio conformance) for lightweight geometry-aware feedback across rounds. IterVision replaces the analytic scorer with a VLM semantic critic (Qwen2.5-VL-72B) that evaluates rendered views via chain-of-thought visual reasoning, assessing spatial coherence and design intent. On a benchmark spanning four canonical geometry families (plates with holes and bolt circles, multi-feature boxes, flanged cylinders, and L-brackets), we evaluate seven foundation models: DeepSeek-V3.2, Qwen3-235B-A22B, Llama-3.3-70B, Gemma-3-27B, GLM-4.5, MiniMax-M2.1, and INTELLECT. Under IterTracer, the four highest-ranked models form a tight cluster (overall mean in [0.885, 0.890]) with 98.97% mesh success, showing that compact instruction-tuned models can match substantially larger systems. VLM-based critique in IterVision yields 100% watertight mesh generation on the leading model while surfacing systematic difficulty on rotationally symmetric geometries such as cylinders, where visual and semantic scoring diverge most. We discuss benchmark design, failure modes, CAD-oriented prompting, and implications for industrial workflows and scalable automated mechanical design.
MecEng, a fully automated benchmark that evaluates LLMs on the creation of multibody simulation models from parameterized textual descriptions, indicates rapidly improving, but still error-prone, mechanical engineering awareness of current LLMs.
Johannes Gerstmayr, Sebastian Weyrer, Tobias Möltner et al.· 0 citations
Recent vision-language models (VLMs) can generate executable CAD programs from images, but existing methods mainly target coarse, general-purpose 3D objects and rarely address the fine-grained geometry and millimeter-level tolerances required in industrial mechanical design. We introduce OmniMech, the first million-scale benchmark for evaluating VLMs on executable CAD generation from industrial manufacturing data. OmniMech contains more than 251,000 fully dimensioned and toleranced 2D orthographic drawings, paired with native CAD models, multi-view renderings, mesh, STEP and B-rep representations, and rich semantic annotations. The benchmark includes four tasks: (1) parametric CAD program synthesis from engineering drawings; (2) diagram-to-3D reasoning for geometrically and structurally consistent reconstruction; (3) annotation-grounded reasoning over dimensions, symbols, feature callouts, and manufacturing constraints; and (4) tool-augmented agentic reasoning using visualization, measurement, CAD execution, and verification tools. Experiments show that current VLMs and CAD-specialized models still struggle with executable program synthesis, fine-grained 3D reconstruction, and reliable enforcement of dimensions and tolerances. We will release the benchmark data, evaluation code, and tool interfaces to support future research.
Taiting Lu, Runze Liu, Ziwei Dong et al.· 0 citations
Advances in large language models (LLMs) have sparked interest in automating parametric CAD modeling through natural language. Existing LLM-based approaches often treat CAD modeling as flat text generation, overlooking the hierarchical structure and geometric constraints inherent in CAD programs. We present CAD-Factory, a Text-to-CAD generation system for language-driven CAD modeling that explicitly models the structural and parametric semantics of CAD programs. Our core contribution is a new formulation of CAD generation as structured program synthesis, coupled with a learnable hierarchical CAD program representation that disentangles structural topology from parameter instantiation. Building on this formulation, CAD-Factory adopts a manager–programmer–reviewer architecture: a planning agent infers program hierarchy, a coding agent instantiates symbolic and numerical parameters, and an evaluation module enforces structural validity and geometric feasibility, which supports structure-aware reasoning, constraint-consistent generation, and interpretable program synthesis. An editor–viewer loop further enables part-aware code refinement through visual feedback, supporting iterative and controllable design workflows. We also contribute a process-oriented annotation pipeline and a Text–CAD dataset with parameter-free or partially specified descriptions that reflect real-world design expression. Extensive experiments demonstrate that CAD-Factory significantly improves structural correctness and geometric consistency across diverse CAD code generation and editing tasks, establishing a structured, interpretable, and robust framework for AI-assisted design.
Yang Liu, Daxuan Ren, Yijie Ding et al.· International Conference on...· 0 citations
Generative models have recently made significant progress in generating 3D content from image or text data, yet existing methods mostly rely on implicit or discretized geometric representations that lack parametric structure or design intent. These limitations make them poorly suited for computer-aided design (CAD). In contrast, code-based CAD representations offer parametric structure and a human-readable presentation but remain underexplored due to the scarcity of datasets. We present a dataset of over 95,000 native OpenSCAD models, primarily containing mechanical parts and engineering components, designed specifically for the programmatic generation of 3D models. The dataset consists of user-created OpenSCAD code, featuring both constructive solid geometry and sketch-and-extrude modeling, along with descriptive metadata, rendered images, and detailed license information. A five-stage pipeline ensures syntactic validity, semantic richness, de-duplication, and quality annotations. To test the dataset's potential, we fine-tuned code-generation large language models for text-to-CAD generation. The evaluation results show consistent improvements in compilation validity and geometric distance to ground-truth samples. By releasing this dataset publicly, we aim to open new opportunities for programmatic 3D model generation.
D. Fresacher, Klaus Diepold· Proceedings of the 32nd ACM...· 0 citations
Automated evaluation is essential for scaling generative 3D systems, where exhaustive human review is costly and slow. However, the reliability of an automated judge depends on the entire evaluation pipeline, not only the underlying vision-language model (VLM), but also how assets are rendered, what visual evidence is provided, how the task is specified, and how human reference labels are constructed. We introduce 3D-DefectBench, a benchmark and framework for systematic analysis of VLM-based 3D defect detection pipelines. It complements holistic ratings and pairwise preferences with nine fine-grained binary defects spanning geometry, texture, and prompt adherence, providing actionable diagnostics for generator development and judge evaluation. Using a balanced factorial design, we vary four pipeline factors, VLM, camera protocol, visual input, and prompt schema, across 84 inference designs and approximately 3.2 million scored defect decisions, followed by staged validation on a broader set of frontier models. Model choice is the largest determinant of agreement with human labels, but the remaining factors also affect performance, interact with model selection, and can change the best configuration. Within the evaluated design space, a compact six-view RGB protocol performs comparably to denser multi-view settings and inputs augmented with depth or surface normals, making it a strong cost-effective default. Under this standardized pipeline, the best of 12 VLM judges still lag behind trained human labelers, while texture agreement drops sharply when expert-consensus labels are replaced by noisier silver labels. These findings show that automated judges should be evaluated as complete pipelines and calibrated across human reference regimes, rather than benchmarked only as standalone models. We release labels, prompts, predictions, and Croissant metadata on Hugging Face.
Zhenyu Zhao, Nanshan Jia, Jihyeon Je et al.· 0 citations
This work presents StructuredEdit, a pipeline that reframes design editing as parameter manipulation rather than pixel generation and embeds hard design constraints into vision-language model fine-tuning by backpropagating pixel-level constraint violations through a lightweight differentiable rasterizer.