infery
← All models

Triposplat

triposplat

3D generationby Tripo3d

TripoSplat is an open-source model from TripoAI / VAST AI Research that converts a single 2D image into high-quality 3D Gaussians using a novel learned density-control approach

Details

Accepts
image

Pricing

Price
6.25 cr / request

Prices in credits (1 credit = $0.01).

Data schema

Input

FieldTypeDescription
seedRandom seed for reproducibility. If None, a random seed is generated.
image_urlrequiredstringURL of the input image to convert into a 3D Gaussian splat.
num_gaussiansintegerTarget number of Gaussians in the output splat. Higher means more detail but larger files and slower rendering. Rounded to the nearest multiple of 32 internally.
output_formatstringenum: ply, splatFile format for the generated Gaussian splat.
guidance_scalenumberClassifier-free guidance strength. Values at or below 1.0 disable guidance; higher values increase detail and adherence to the input image, but too high can oversaturate colors.
num_inference_stepsintegerNumber of flow-matching sampler steps. More steps improve fidelity with a roughly linear runtime cost.
enable_safety_checkerbooleanEnable safety checking of the input image.

Output

FieldTypeDescription
seedintegerThe seed used for generation.
timingsobjectPer-stage timings in seconds.
model_meshGenerated 3D Gaussian splat file (.ply or .splat).
num_gaussiansintegerThe actual number of Gaussians produced.
preprocessed_imageThe background-removed RGB composite the model actually saw, rendered on a black background.