arXiv cs.AI· Nianchen Deng, Jiaxin Ai, Tao Hu, Shu Zou, Yurui Dong, Siqi Li, Xinyu Cai, Xuemeng Yang, Licheng Wen, Hongbin Zhou, Hairong Zhang, Pinlong Cai, Botian Shi·· 12 小时前AI 评分38
IndustryForge-27B：面向工业 CAD 的领域增强多模态基础模型
IndustryForge-27B: A Domain-Enhanced Multimodal Foundation Model for Industrial CAD
AI 导读
基于 Qwen3.5-VL-27B 构建的 IndustryForge-27B 通过整合约 52k 多模态样本的六个工业 CAD 子语料并采用统一多任务 SFT 训练，在四个 CAD 领域基准上平均较基座模型提升 33.65 个百分点，且四项均超过闭源模型 GPT-5.4。
正文 · 原文
Authors:Nianchen Deng, Jiaxin Ai, Tao Hu, Shu Zou, Yurui Dong, Siqi Li, Xinyu Cai, Xuemeng Yang, Licheng Wen, Hongbin Zhou, Hairong Zhang, Pinlong Cai, Botian Shi
Abstract:Automating industrial CAD design and manufacturing places distinctive demands on multimodal foundation models: the model must see engineering drawings and 3D geometry screenshots, write correct parametric-modelling scripts and Windows COM API code, and cover the full range from single parts to assemblies. General-purpose multimodal models fall short on these tasks, while single-task fine-tuning is too narrow to support the diverse calls that upper-layer agents issue. We build IndustryForge-27B on top of Qwen3.5-VL-27B by curating and integrating six industrial-CAD sub-corpora totalling $\sim$52k multimodal samples---CAD Visual QA (CAD-VQA), parametric CAD code (text2cadquery), assembly-level CAD code (text2cadquery-assembly), and three COM sub-corpora for Inventor / SolidWorks (com_2d / com_3d / com_assembly)---and training with a unified multi-task SFT recipe. Across four CAD-domain benchmarks IndustryForge-27B lifts the base model by $+33.65$~pp on average and outperforms the strong closed-source model GPT-5.4 on all four; across eleven general-capability benchmarks it retains, and slightly improves upon, the base model ($+1.56$~pp mean, no catastrophic forgetting). IndustryForge-27B will serve as the common substrate for downstream industrial-agent projects, providing a unified starting point for a full-stack industrial agent that spans from CAD design to industrial-software operation, from parts to assemblies, and from single-shot generation to closed-loop self-improvement. Huggingface model: this https URL
| Comments: | 14 pages |
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2607.28050 [cs.AI] |
| (or arXiv:2607.28050v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2607.28050 arXiv-issued DOI via DataCite |
Submission history
From: Nianchen Deng [view email]
[v1]
Thu, 30 Jul 2026 11:28:21 UTC (4,964 KB)
[v2]
Tue, 20 Oct 2026 08:03:48 UTC (4,977 KB)
来源：arXiv cs.AI · arxiv.org