Open-source Idol model turns a single image into 3D in 45s on 8GB VRAM
Promptmethus · x · 2026-11-04
Comfy Org released Idol, an Apache-2.0 image-to-3D model fine-tuned from TripoSplat, on Hugging Face, and the author calls it a game-changer for consumer hardware.
- Low hardware bar: runs smoothly on just 8GB VRAM, versus the 24GB+ typically required for local 3D generation.
- Fast iteration: a single image becomes a full 3D model in about 45 seconds, enabling rapid sketch-and-retry workflows.
- Setup: uses the Image to Gaussian Splat workflow in ComfyUI, with DINOv3 and FLUX2-VAE components.
- Caveat: exported Splats have niche topology and can't directly replace high-precision meshes, but they're great for quick scene assets, VRChat character prototyping, and indie dev work.
Weights are on Hugging Face (Comfy-Org/Idol), with 6,600 downloads last month.
More from Multimodal
- Experimental video mixes visual styles, generated with Seedance 2.5 on Adobe Firefly — umesh_ai · 2026-11-04
- Hailuo AI claims H3 scored perfect on language tests; Japanese creator praises its Japanese signage fidelity — Hailuo_AI · 2026-11-04
- YANAMI: 400K-dialog multimodal dataset shows models still fail at stateful multiturn interaction — Shengyu Wang · 2026-11-04
- ComfyUI v0.44.0 released with new Patina texture nodes — Fcking_Chuck · 2026-11-04
- Krea 3 technical report released: a deep dive into data, architecture and training — bdsqlsz · 2026-11-04
- 22 Creative SAM 3.1 Use Cases From Meta's AI Developer Hackathon, From Wildfire Detection to Document Reassembly — Vjeux · 2026-11-04