A2Z-10M+: Geometric Deep Learning with A-to-Z BRep Annotations for AI-Assisted CAD Modeling and Reverse Engineering

Pritham Kumar Jena; Bhavika Baburaj; Tushar Anand; Vedant Dutta; Vineeth Ulavala; Sk Aziz Ali

cs.CVMar 13, 2026

A2Z-10M+: Geometric Deep Learning with A-to-Z BRep Annotations for AI-Assisted CAD Modeling and Reverse Engineering

Authors: Pritham Kumar Jena, Bhavika Baburaj, Tushar Anand, Vedant Dutta, Vineeth Ulavala, Sk Aziz Ali

Organizations: 3D Vision Group (3DVG) · BITS Pilani, Hyderabad, India

Abstract

Reverse engineering and rapid prototyping of computer-aided design (CAD) models from 3D scans, sketches, or simple text prompts are vital in industrial product design. However, recent advances in geometric deep learning techniques lack a multi-modal understanding of parametric CAD features stored in their boundary representation (BRep). This study presents the largest compilation of 10 million multi-modal annotations and metadata for 1 million ABC CAD models, namely A2Z, to unlock an unprecedented level of BRep learning. A2Z comprises (i) high-resolution meshes with salient 3D scanning features, (ii) 3D hand-drawn sketches equipped with (iii) geometric and topological information about BRep co-edges, corners, and surfaces, and (iv) textual captions and tags describing the product in the mechanical world. Creating such carefully structured, large-scale data, which requires nearly 5 terabytes of storage to leverage unparalleled CAD learning/retrieval tasks, is very challenging. The scale, quality, and diversity of our multi-modal annotations are assessed using novel metrics, GPT-5, Gemini, and extensive human feedback mechanisms. To this end, we also merge an additional 25,000 CAD models of electronic enclosures (e.g., tablets, ports) designed by skilled professionals with our A2Z dataset. Subsequently, we train and benchmark a foundation model on a subset of 150K CAD models to detect BRep co-edges and corner vertices from 3D scans, a key downstream task in CAD reverse engineering. The annotated dataset, metrics, and checkpoints will be publicly released to support numerous research directions.

Figures & tables

Dataset	CAD Models	#Models	BRep Labels	Text Caption	3D Scans	3D Sketch	Design History
DeepCAD [ 67 ]	Simple ✓	170,000+	✗	✗	✗	✗	✓
Fusion360 [ 65 ]	Simple ✓	$\sim$ 8,000	✗	✗	✗	✗	✓
CADParser [ 73 ]	Simple ✓	41,000	✗	✗	✗	✗	✓
CAD-MLLM	Simple ✓	453,220	✗	✓	✗	✗	✓
ABC [ 29 ]	Cmplx. ✓	1,000,000	✗	✗	✗	✗	✗
CC3D-Ops [ 12 ]	Cmplx. ✓	$\sim$ 50,000	✓	✗	✓	✗	✗

Table 1: Comparison of open-source CAD datasets w.r.t A2Z

Figure 2: Low-poly CAD meshes are compared to high-quality scans after a series of geometry processing steps ( Sec. 3.1 ).

Figure 3: BRep annotations on 3D scans of A2Z dataset.

Figure 4: Data and annotation analytics of A2Z

Figure 5: Five million sketches with BRep annotations

Figure 6: $\sim$ 5K models of charging points, serial bus etc.

Method	Sketch Artist Level	Gemini	GPT	Human
Face IDs	—	8.43	8.79	8.37
Face Type	—	7.88	8.71	8.05
Sketches	2	8.21	7.47	8.36
	3	7.87	7.71	9.12
	4	7.97	8.11	9.03
	5	8.31	8.43	9.61

Table 2: Visual quality scores (higher is better).

Level MLTD Unigram Bigram Low 59.32 856 11355 High 70.52 908 12181 (a) Avg. MLTD [ 55 ] and uni/bigrams.

Method ID Type Loop Boundary 99.37 97.67 99.99 Face 99.93 99.83 — (b) Annotation coverage (%).

Table 3: Two tables highlights the lexical diversity ( left ) on captions and 3D scans’ annotation coverage statistics ( right ).

Figure 7: Qualitative Comparison of state-of-the-art methods in BRep Boundary and Junction classification as a downstream task.

Figure 8: Realism of Sketch.: We compute a CLIP cosine-similarity score of 30.58 between generated captions and the corresponding 2D sketches, confirming that textual descriptions are well-grounded in visual content. Additionally, a tag-level similarity score of 64.7% (CAD-derived tags vs. sketch-derived tags) demonstrates that the hierarchical tag vocabulary remains consistent across the BRep and sketch modalities.

Evaluator	Modality	Level-2	Level-3	Level-4	Level-5
GPT-5	sketch	$7.65\pm 0.34$	$7.20\pm 0.48$	$5.53\pm 0.38$	$5.05\pm 0.39$
Gemini	sketch	$\mathbf{9.01\pm 0.28}$	$7.33\pm 0.52$	$6.06\pm 0.58$	$3.92\pm 0.47$
Human	sketch	$7.20\pm 0.32$	$\mathbf{7.86\pm 0.38}$	$7.85\pm 0.44$	$7.67\pm 0.71$

Table 6: Realism scores (mean

\pm

std, 1–10 scale) for multi-level 2D sketches evaluated by automated VLM raters and human engineering students. Higher is better. Level-3 achieves the strongest evaluator consensus, balancing abstraction with recognizability.

Evaluator	Modality	Step-I	Step-II	Step-III	Step-IV
GPT-5	scan	$6.51\pm 0.84$	$6.54\pm 1.01$	$7.58\pm 0.58$	$\mathbf{9.41\pm 0.60}$
Gemini	scan	$7.73\pm 1.48$	$7.91\pm 0.92$	$8.54\pm 0.78$	$\mathbf{9.28\pm 0.81}$
Human	scan	$6.32\pm 0.29$	$6.58\pm 0.80$	$7.20\pm 0.51$	$\mathbf{9.83\pm 0.22}$

Table 7: Realism scores (mean

\pm

std, 1–10 scale) for synthesized scans at each pipeline stage, evaluated against Artec3D ground-truth scans of CC3D models. Refer to Figure. 12 for visuals. Step IV achieves near-perfect human consensus.

Figure 9: Hierarchical categorization of 1 million CAD models for retrieval and causal mining. The two examples illustrate how – (on the right) one can fetch some CAD models when asked to find – ‘A mechanical component associated to Gear type transmission systems supporting Axial rotation’.

Category	Tag Count	Percentage
Geometric Features	12,230	45.13%
Unclassified	6,068	22.39%
Mechanical Components	4,601	16.98%
Structural Elements	2,724	10.05%
Electronic Components	801	2.96%
Fasteners	678	2.50%

Table 8: Distribution of the

27{,}102

unique tags over top-level categories and their Percentage (

\%

) of CAD models.

Figure 10: This figure visually describes how evaluation and scoring mechanisms from GPT-5, Gemini, and the Human feedback-based evaluation process for different types of annotations on 10K randomly selected samples.

Figure 11: Boundary and Corner detection on randomly selected real-world 3D scan data from the Grab-CAD repository.

Figure 12: Visual comparison between CC3D [ 6 ] CAD models, its real scan pairs, and synthesized scans from CAD (ours)

Figure 13: Visual elaboration of the textual captions on more complex CAD models.

Figure 14: Visual elaboration of the textual captions on more complex CAD models

Figure 15: Visual elaboration of the textual captions on more complex CAD models

Figure 16: Visuals of some randomly selected CAD models of charging ports under A2Z. Color codes define face-Types.

Figure 17: Visuals of same CAD models of charging ports under A2Z where color codes define face-IDs.

Figure 18: Visuals of some randomly selected samples of CAD models for Tablet type enclosures (from Approx. 20K Electronic enclosure models) with varying corner geometry and topology ( at bottom ). A closer look at the BRep faces, edges, and vertices near the corner parts ( on top ) shows how many faces are adjacent to each edge.

Figure 19: More visual examples of BRep Annotations for reverse engineering from 3D scans/sketches.

Chunk	Boundary		Junction
	Recall	Precision	Recall	Precision
0	0.972	0.885	0.690	0.740
1	0.972	0.905	0.710	0.774
2	0.980	0.883	0.681	0.703
3	0.977	0.883	0.688	0.739
4	0.964	0.841	0.698	0.701

Table 9: Precision-recall measures for all 100 chunks of A2Z using our foundation model.

Boundary Representation Modeling

See all 9 papers

DreamCAD: Scaling Multi-modal CAD Generation using Differentiable Parametric Surfaces
Mar 5, 2026Mohammad Sadil Khan, Muhammad Usama, Rolandos Alexandros Potamias +4Text-to-CAD Generation Parametric CAD Modeling
DFKI, Germany · RPTU Kaiserslautern, Germany · Imperial College London, United Kingdom +1
BrepCoder: A Unified Multimodal Large Language Model for Multi-task B-rep Reasoning
Feb 25, 2026Mingi Kim, Yongjun Kim, Jungwoo Kang +1Multimodal Large Language Models
Chungnam National University, Daejeon, Republic of Korea
BRepCLIP: Contrastive Multimodal Pretraining on BRep Primitives for CAD Understanding
Jun 3, 2026Muhammad Usama, Didier Stricker, Mohammad Sadil Khan +13D Shape Retrieval Multimodal Pretraining
1DFKI, Germany · 2RPTU Kaiserslautern-Landau, Germany

Mar 5, 2026J/K move · Enter open · S save

DreamCAD: Scaling Multi-modal CAD Generation using Differentiable Parametric Surfaces

Mohammad Sadil Khan, Muhammad Usama, Rolandos Alexandros Potamias, Didier Stricker+3

DFKI, Germany · RPTU Kaiserslautern, Germany · Imperial College London, United Kingdom+1

Text-to-CAD Generation Parametric CAD Modeling

Open paper AskPDF arXiv

Computer-Aided Design (CAD) relies on structured and editable geometric representations, yet existing generative methods are constrained by small annotated datasets with explicit design histories or boundary representation (BRep) labels. Meanwhile, millions of unannotated 3D meshes remain untapped, limiting progress in scalable CAD generation. To address this, we propose DreamCAD, a multi-modal generative framework that directly produces editable BReps from point-level supervision, without CAD-specific annotations. DreamCAD represents each BRep as a set of parametric patches (e.g., Bézier surfaces) and uses a differentiable tessellation method to generate meshes. This enables large-scale training on 3D datasets while reconstructing connected and editable surfaces. Furthermore, we introduce CADCap-1M, the largest CAD captioning dataset to date, with 1M+ descriptions generated using GPT-5 for advancing text-to-CAD research. DreamCAD achieves state-of-the-art performance on ABC and Objaverse benchmarks across text, image, and point modalities, improving geometric fidelity and surpassing 75% user preference. Code and dataset will be publicly available.

Abstract

Boundary Representation Modeling

Explore similar work

DreamCAD: Scaling Multi-modal CAD Generation using Differentiable Parametric Surfaces

BrepCoder: A Unified Multimodal Large Language Model for Multi-task B-rep Reasoning

BRepCLIP: Contrastive Multimodal Pretraining on BRep Primitives for CAD Understanding