Note: This thing kinda sucks compared to trellis. The workflow sat in my drafts for a while but I figured I better get it out there because somebody might find it useful. I probably wont provide much support for this one unless I'm doing something dumb in the workflow or if there is a way to make this thing better. Type the name of any object in a picture and it turns into 3D. You get a .ply file you can turn around, and a spin video if you want one.Most picture-to-3D workflows want a clean product shot. This one starts from any photo: type the thing you want, like <code>the camera</code>, and only that goes to Idol.<h3 id="how-to-use-it">How to use it</h3><ol><li>Load your picture.</li><li>In Type what you want, type the thing to take out, like <code>the camera</code>.</li><li>Press Run. What it found shows what was picked.</li><li>Drag inside Save Splat to turn your object around.</li></ol><h3 id="switches">Switches</h3><ul><li>Fill in hidden parts draws what a hand covers. Off at the start. Leave it off for people and statues. Big download, about 17 GB.</li><li>Sharpen with SeedVR2 adds detail before the build. On at the start.</li><li>1 million splats for a denser, bigger file.</li><li>Looser framing if the edges get cut off.</li><li>Save a spin video makes a 3 second loop, seen straight on at eye level. Click background to pick its color.</li><li>Make a mesh (GLB) for Blender or a game engine.</li></ul><h3 id="what-you-need">What you need</h3><ul><li>AusBoss nodes 2.21.2 or newer (update first), ComfyUI 0.43.2 or newer. No other node packs.</li><li><a target="_blank" rel="ugc" href="https://huggingface.co/Comfy-Org/Idol/resolve/main/diffusion_models/idol_fp16.safetensors">idol_fp16.safetensors</a> → <code>models/diffusion_models</code> · 741 MB</li><li><a target="_blank" rel="ugc" href="https://huggingface.co/Comfy-Org/Idol/resolve/main/vae/idol_decoder_fp16.safetensors">idol_decoder_fp16.safetensors</a> → <code>models/vae</code> · 576 MB</li><li><a target="_blank" rel="ugc" href="https://huggingface.co/Comfy-Org/Idol/resolve/main/clip_vision/dino_v3_vit_h_bf16.safetensors">dino_v3_vit_h_bf16.safetensors</a> → <code>models/clip_vision</code> · 1.68 GB</li><li><a target="_blank" rel="ugc" href="https://huggingface.co/Comfy-Org/Idol/resolve/main/vae/flux2-vae.safetensors">flux2-vae.safetensors</a> → <code>models/vae</code> · 336 MB</li><li><a target="_blank" rel="ugc" href="https://huggingface.co/Comfy-Org/sam3.1/resolve/main/checkpoints/sam3.1_multiplex_fp16.safetensors">sam3.1_multiplex_fp16.safetensors</a> → <code>models/checkpoints</code> · 1.75 GB</li><li>For Sharpen with SeedVR2 (on at the start): <a target="_blank" rel="ugc" href="https://huggingface.co/Comfy-Org/SeedVR2/resolve/main/diffusion_models/seedvr2_3b_int8_convrot.safetensors">seedvr2_3b_int8_convrot.safetensors</a> → <code>models/diffusion_models</code> · 3.46 GB</li><li>For Sharpen with SeedVR2: <a target="_blank" rel="ugc" href="https://huggingface.co/Comfy-Org/SeedVR2/resolve/main/vae/seedvr2_ema_vae_fp16.safetensors">seedvr2_ema_vae_fp16.safetensors</a> → <code>models/vae</code> · 501 MB</li><li>Only for Fill in hidden parts (off at the start, about 17 GB in all): <a target="_blank" rel="ugc" href="https://huggingface.co/Comfy-Org/Qwen-Image-2.1/resolve/main/diffusion_models/qwen_image_2.1_int8_convrot.safetensors">qwen_image_2.1_int8_convrot.safetensors</a> → <code>models/diffusion_models</code> · 7.26 GB</li><li>Only for Fill in hidden parts: <a target="_blank" rel="ugc" href="https://huggingface.co/Comfy-Org/Qwen-Image-2.1/resolve/main/text_encoders/qwen3vl_8b_int8_convrot.safetensors">qwen3vl_8b_int8_convrot.safetensors</a> → <code>models/text_encoders</code> · 9.35 GB</li><li>Only for Fill in hidden parts: <a target="_blank" rel="ugc" href="https://huggingface.co/Comfy-Org/Qwen-Image-2.1/resolve/main/vae/qwen_image_2.1_vae_bf16.safetensors">qwen_image_2.1_vae_bf16.safetensors</a> → <code>models/vae</code> · 676 MB</li><li>Only for Fill in hidden parts, rename <code>model.safetensors</code> to this name: <a target="_blank" rel="ugc" href="https://modelscope.ai/models/DiffSynth-Studio/Qwen-Image-2.1-LayerExtract/resolve/master/model.safetensors">Qwen-Image-2.1-LayerExtract.safetensors</a> → <code>models/loras</code> · 168 MB</li></ul><h3 id="tips">Tips</h3><ul><li>Works best on a clear, front-on photo of one object, in even light.</li><li>Idol guesses the depth from one picture, so some shapes come out squashed. Try another photo.</li><li>Say it more clearly if it picks the wrong thing: <code>the red umbrella</code>, not <code>umbrella</code>.</li><li>The back and sides are made up to match the front. Change the Seed and run again for a different guess.</li><li>It is built for one object. People and rooms don't come out well.</li><li>In the 3D viewer, scroll to zoom in. It opens small and far away.</li></ul><h3 id="credits">Credits</h3>Idol is by speridlabs (Apache-2.0), with ComfyUI files from Comfy-Org. DINOv3 and SAM 3 are by Meta, the FLUX.2 VAE by Black Forest Labs, SeedVR2 by ByteDance Seed, Qwen Image 2.1 by Qwen, and Layer Extract by DiffSynth-Studio. Each has its own license. Built with DINOv3.v1.0 (2026-11-03): first release.Questions and bugs: <a target="_blank" rel="ugc" href="https://github.com/ausboss/ComfyUI-AusBoss/issues">GitHub issues</a>. Videos on YouTube @ausbossdev. More from me on <a target="_blank" rel="ugc" href="https://x.com/Zanzibased">X @Zanzibased</a> and <a target="_blank" rel="ugc" href="https://github.com/ausboss/ComfyUI-AusBoss">GitHub</a>.

Idol Picture to 3D | Type an Object, Get It in 3D

First release. Needs ComfyUI 0.43.2 or newer and AusBoss nodes 2.21.2 or newer.

Files with SHA256: d124d5d2c4d67a429b4d58feb02590e18aed59cfa3a5f4e0a53583c18070c7f6