Search papers, labs, and topics across Lattice.
BlenderRAG is introduced, a retrieval-augmented generation system that synthesizes Blender code from natural language descriptions to generate 3D objects. It leverages a multimodal dataset of 500 expert-validated examples to retrieve semantically similar examples during code generation, addressing the challenges of syntactic errors and geometric inconsistencies common in LLM-generated code. BlenderRAG significantly improves compilation success rates and semantic alignment, demonstrating a practical approach to high-fidelity 3D object generation.
LLMs can now generate 70% syntactically correct and geometrically consistent 3D objects from text, thanks to retrieval-augmented code synthesis.
Automatic generation of executable Blender code from natural language remains challenging, with state-of-the-art LLMs producing frequent syntactic errors and geometrically inconsistent objects. We present BlenderRAG, a retrieval-augmented generation system that operates on a curated multimodal dataset of 500 expert-validated examples (text, code, image) across 50 object categories. By retrieving semantically similar examples during generation, BlenderRAG improves compilation success rates from 40.8% to 70.0% and semantic normalized alignment from 0.41 to 0.77 (CLIP similarity) across four state-of-the-art LLMs, without requiring fine-tuning or specialized hardware, making it immediately accessible for deployment. The dataset and code will be available at https://github.com/MaxRondelli/BlenderRAG.
