Diffusion model from Speridlabs that turns a single image into a 3D Gaussian splat object at adjustable detail levels.
Updated EU status checked Last shown
Sources: the vendor's own pages (linked above) and creator videos at the linked timestamps. EU status is information, not legal advice; see how we check.
Idol is a 0.66B-parameter diffusion model from Speridlabs Research that converts one image into a 3D object represented as Gaussians (a Gaussian splat). The input can be a photograph, a render, or an image generated by Speridlabs' Iris-3B model. The output can be viewed from all sides, including angles the input image did not show.
Objects can be generated at several levels of detail, from 32,768 to 1,048,576 Gaussians. The model runs in under 8 GB of GPU memory. According to the project page, one image becomes 262,144 Gaussians in 3.4 s on a B300 and 6.7 s on an H100 at 50 steps. The same task takes 19.1 s on an RTX 4090 and 46.2 s on an RTX 3080.
The project page reports a benchmark against TripoSplat, TRELLIS and LGM on one H100. Idol scores best on most metrics, including LPIPS, IoU, DINOv2 and sharpness distance. LGM has the highest input-view PSNR and the fastest request time. A YouTube review says code and models have been released, but no repository, weights link or license appeared in the material provided, so these details are unconfirmed.
Watch the moment: the exact second top creators show it. Videos load only when you press play, via youtube-nocookie.com.
Turns a single image into a 3D object using Gaussian splatting, with adjustable quality. Only 0.66B parameters, runs on 8GB VRAM, about 19 s on a 4090. Code and models released.
Closest neighbours by meaning, not just by category.
Model · Sperid Labs
Open 3B-parameter text-to-image flow-matching model that generates directly in pixel space, without a VAE.
Model
Feed-forward 3D Gaussian Splatting model that reconstructs scenes from arbitrary input views, with or without camera poses.
Model · Princeton
Fully open 3B text-to-image diffusion model from Princeton with released weights, training code, and data pipeline.
Model · Nvidia
NVIDIA research model that turns a single image into a queryable 3D Gaussian scene with semantics and object instances.
Model
Pixel-space text-to-image diffusion model transferred from latent diffusion models, without a VAE, supporting 4K output.
Model
Research model that unifies 3D scene generation and reconstruction in a single pixel-space diffusion model.
