mlsanityData to 14 Oct 2026

BlenderLLM: Training Large Language Models for Computer-Aided Design with Self-improvement

arXiv 2412.14203 ↗First posted 16 Dec 2024cs.HC cs.AIcited by 29

On the map: Synthetic data and self-improvement for LLMs → Self-rewarding alignment and self-training

Abstract (arXiv)

The application of Large Language Models (LLMs) in Computer-Aided Design (CAD) remains an underexplored area, despite their remarkable advancements in other domains. In this paper, we present BlenderLLM, a novel framework for training LLMs specifically for CAD tasks leveraging a self-improvement methodology. To support this, we developed a bespoke training dataset, BlendNet, and introduced a comprehensive evaluation suite, CADBench. Our results reveal that existing models demonstrate significant limitations in generating accurate CAD scripts. However, through minimal instruction-based fine-tuning and iterative self-improvement, BlenderLLM significantly surpasses these models in both functionality and accuracy of CAD script generation. This research establishes a strong foundation for the application of LLMs in CAD while demonstrating the transformative potential of self-improving models in advancing CAD automation. We encourage further exploration and adoption of these methodologies to drive innovation in the field. The dataset, model, benchmark, and source code are publicly available at https://github.com/FreedomIntelligence/BlenderLLM

Key facts. Cited by 29 papers in the map (data through 14 Oct 2026). Placed in Self-rewarding alignment and self-training, one of 13 directions in Synthetic data and self-improvement for LLMs. Builds on 15 of its 32 references that are in the map. First posted on arXiv 16 Dec 2024.

Builds on

15 of 32 references.

References · 15

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2023GPT-4 Technical Reportcited by 18,396
  3. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,599
  4. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 14,132
  5. 2024Gemini 1.5: Unlocking multimodal understanding across millions of tokens of contextcited by 3,224
  6. 2022Self-Instruct: Aligning Language Models with Self-Generated Instructionscited by 2,724
  7. 2023Code Llama: Open Foundation Models for Codecited by 2,139
  8. 2024Mixtral of Expertscited by 1,823
  9. 2024Qwen2.5-Coder Technical Reportcited by 1,203
  10. 2024DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Modelcited by 1,135
  11. 2024CadVLM: Bridging Language and Vision in the Generation of Parametric CAD Sketchescited by 28
  12. 2023MLLM-Bench: Evaluating Multimodal LLMs with Per-sample Criteriacited by 27
3 more
  1. 20243D-PreMise: Can Large Language Models Generate 3D Shapes with Sharp Features and Parametric Control?cited by 21
  2. 2020Sketch2CAD: Sequential CAD Modeling by Sketching in Contextcited by 14
  3. 2024CADgpt: Harnessing Natural Language Processing for 3D Modelling to Enhance Computer-Aided Design Workflowscited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 9; 2026: 20.

29 papers.

Sort:

By area:

  1. 2025CAD-Coder: An Open-Source Vision-Language Model for Computer-Aided Design Code Generationcited by 43
  2. 2025Large Language Models for Computer-Aided Design: A Surveycited by 18
  3. 20263DCodeBench: Benchmarking Agentic Procedural 3D Modeling Via Codecited by 15
  4. 2026CADBench: A Multimodal Benchmark for AI-Assisted CAD Program Generationcited by 11
  5. 2026SceneCode: Executable World Programs for Editable Indoor Scenes with Articulated Objectscited by 11
  6. 2025Automated CAD Modeling Sequence Generation from Text Descriptions via Transformer-Based Large Language Modelscited by 8
  7. 2026VoxelCodeBench: Benchmarking 3D World Modeling Through Code Generationcited by 8
  8. 2026A Survey on Evaluating Quality and Trustworthiness in LLM-Generated Datacited by 6
  9. 2026TOOLCAD: Exploring Tool-Using Large Language Models in Text-to-CAD Generation with Reinforcement Learningcited by 6
  10. 2025ShapeLib: Designing a library of programmatic 3D shape abstractions with Large Language Modelscited by 5
  11. 2025ToolGrad: Efficient Tool-use Dataset Generation with Textual "Gradients"cited by 4
  12. 2025ShapeCraft: LLM Agents for Structured, Textured and Interactive 3D Modelingcited by 3
17 more
  1. 2026Nova3D: Code-Native Generation of Programmable 3D Assetscited by 2
  2. 20253DFroMLLM: 3D Prototype Generation only from Pretrained Multimodal LLMscited by 1
  3. 2025BrepLLM: Enabling Large Language Models to Understand Boundary Representationscited by 1
  4. 2026From Idea to Co-Creation: A Planner-Actor-Critic Framework for Agent Augmented 3D Modelingcited by 1
  5. 2026Aggregating LLM-Based Weak Verifiers for Spatial Layout Generationcited by 1
  6. 20263D-CoS: A New 3D Reconstruction Paradigm Based on VLM Code Synthesiscited by 1
  7. 2026BIM-Edit: Benchmarking Large Language Models for IFC-Based Building Information Modelingcited by 1
  8. 2026aDSL: Agentic 3D Creation via Joint Agent-Program Designcited by 1
  9. 2026ViSculpt: Visual-Centric Agentic Geometry Editingcited by 1
  10. 20263DHarnessBench: Probing Agentic 3D-to-Code Capabilities of Frontier Vision-Language Modelscited by 1
  11. 2025FloraForge: LLM-Assisted Procedural Generation of Editable and Analysis-Ready 3D Plant Geometric Models For Agricultural Applicationscited by 0
  12. 2026See2Refine: Vision-Language Feedback Improves LLM-Based eHMI Action Designerscited by 0
  13. 2026Bioinspired123D: Generative 3D Modeling System for Bioinspired Structurescited by 0
  14. 2026Learning Hierarchical and Geometry-Aware Graph Representations for Text-to-CADcited by 0
  15. 2026BlenderRAG: High-Fidelity 3D Object Generation via Retrieval-Augmented Code Synthesiscited by 0
  16. 2026PhysCodeGym: Benchmarking Coding Agents for Physically Verifiable 4D Reconstructioncited by 0
  17. 2026Code2Worlds-2: Physically Plausible Agent Harness for Embodied World Simulationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025CAD-Editor: A Locate-then-Infill Framework with Automated Training Data Synthesis for Text-Based CAD Editing
  2. 2025A Solver-Aided Hierarchical Language for LLM-Driven CAD Design
  3. 2025Text-to-CadQuery: A New Paradigm for CAD Generation with Scalable Large Model Capabilities
  4. 2025cadrille: Multi-modal CAD Reconstruction with Reinforcement Learning
  5. 2026Foundation Models for Automatic CAD Generation
  6. 2026LLM4CAD-Editor: An Intent-Aware Large Language Model Framework for Multi-Level Computer-Aided Design Editing ↗
  7. 2024Generating CAD Code with Vision-Language Models for 3D Designs
  8. 2025Seek-CAD: A Self-refined Generative Modeling for 3D Parametric CAD Using Local Inference via DeepSeek