Amazon product photo to CAD model in 6m15s with an LLM
_Stocko_ · x · 2026-09-20
Stocko demoed a counterintuitive workflow: dropping a plain Amazon product image into an LLM and getting a CAD model of the item in just 6 minutes 15 seconds. It's a striking example of multimodal models doing practical 3D/industrial design work straight from an ordinary e-commerce photo — no manual modeling software in the loop.
More from Multimodal
- Real-time generation demo shows hybrid image model generating whole-image codes first — ostrisai · 2026-09-20
- Ostris shows DiT code-to-image reproduction progress in AR-diffusion hybrid trained on one RTX 6000 Pro — ostrisai · 2026-09-20
- First samples from Ostris's AR-diffusion hybrid show Qwen3-VL-4B grasping prompt-to-image generation — ostrisai · 2026-09-20
- AI Toolkit author trains AR-diffusion hybrid image model in 24 hours on a single RTX 6000 Pro — ostrisai · 2026-09-20
- Voice-first Gemini Live demo: real-time transcription, reasoning and tool calling — AI_Andrew · 2026-09-20
- YuE2 cover generation: per-song CFG/temp configs that beat the ComfyUI defaults — LilBrownBebeShoes · 2026-09-20