P3D-Bench: Benchmarking MLLMs for Parametric 3D Generation and Structural Reasoning

2026-06-09

AI Summary

Existing benchmarks struggle to systematically evaluate the geometric accuracy, semantic alignment, and structural consistency of multimodal large language models in parametric 3D program generation. To address this gap, this work introduces P3D-Bench, the first comprehensive benchmark specifically designed for parametric 3D programs. It establishes a unified evaluation protocol across three task types—text-to-program, image-to-program, and assembly-to-program—with a multidimensional scoring framework that assesses executability, geometric fidelity, topological correctness, adherence to textual constraints, cross-view semantic alignment, and part-level relational reasoning. Experimental results demonstrate that while current models can capture coarse shape characteristics, they exhibit significant deficiencies in precise parameterization, multi-part assembly, and faithful structural recovery.

Abstract

Multimodal large language models can write code to produce complex programs as well as use programs to do 3D modeling, which opens up a new avenue for 3D generation powered by their priors, world knowledge and reasoning. Yet existing benchmarks rarely evaluate 3D modeling through code. Such modeling demands more than runnable code: from a text or visual specification, a model must generate a parametric 3D program that is geometrically precise, semantically aligned and assembly-consistent. We introduce P3D-Bench, a benchmark for parametric 3D generation. Unlike a 3D mesh, a parametric 3D program exposes explicit dimensions, construction operations and part relations, revealing whether a model recovers a design's structure, not just its appearance. Under a unified protocol, P3D-Bench covers three task families (Text-to-3D, Image-to-3D and Assembly-3D) and scores each output for executability, geometric fidelity, topology, text-grounded constraints, multiview semantic alignment and part-level structure. We evaluate frontier MLLMs and text-only LLMs on 400 text cases, 400 image cases and 203 annotated assemblies, with domain-specific models as reference points. Our extensive evaluation yields three findings. First, assemblies are the hardest setting, where models still fail to compose multiple parts into a coherent structure. Second, models can often recover the global shape and semantic identity of the target object, yet fail to reproduce the precise parametric geometry specified by the input. Third, part-level modeling remains weak on assemblies, where models recover neither the geometry of each part nor the right number of parts. These results position P3D-Bench as a benchmark for evaluating precise parametric geometry and part-level structure in parametric 3D generation.

Problem

• parametric 3D generation
• structural reasoning
• multimodal large language models
• 3D modeling benchmark
• part-level structure

Innovation

• parametric 3D generation
• structural reasoning
• multimodal large language models
• 3D program synthesis
• P3D-Bench