P3D-Bench: Benchmarking MLLMs for Parametric 3D Generation and Structural Reasoning
2026-06-09 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors created P3D-Bench, a test to evaluate how well language models can generate 3D designs using code. Unlike simple 3D shapes, these designs include exact measurements and parts that fit together properly. They tested several advanced models on tasks like making 3D models from text, images, or assembling parts, and found that while models can capture the overall shape and meaning, they struggle with detailed, precise geometry and putting parts together correctly. This benchmark helps measure how good models are at creating accurate and structured 3D programs.
Multimodal large language modelsParametric 3D modelingText-to-3D generationImage-to-3D generationAssembly-3DGeometric fidelityTopologySemantic alignmentPart-level structure
Authors
Yikang Yang, Zhanpeng Hu, Youtian Lin, Mengqi Zhou, Jingxi Xu, Feihu Zhang, Jiaheng Liu, Yao Yao
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.