# VAST-AI-Research/TripoSplat

TripoSplat converts a single 2D image into high-quality and variable number of 3D Gaussians, developed by TripoAI.

Repository: https://github.com/VAST-AI-Research/TripoSplat
Canonical: https://ross.abutalabs.com/products/triposplat
Language: Python
License: MIT
License Family: permissive
Last push: 2026-08-13T09:44:34+00:00

## Health v2 (maintenance only)
Score: 57/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 97, release rhythm 35, longevity 6
- inputs: {"age_days": 93, "days_push": 20, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1232, forks 140 (observed 2026-08-28T04:04:03.725378+00:00)

## What it is
TripoSplat is an inference-only Python library from TripoAI that converts a single 2D image into high-quality 3D Gaussian splats with a variable number of Gaussians (up to 262,144). It ships with minimal, readable code (~2,000 LOC across two files), near-zero dependencies, and official ComfyUI workflow support.

## Use cases
- convert a single image into a 3D gaussian splat model
- generate 3D assets for game development from photos
- create 3D content for AR/VR applications from images
- build simulation environments with image-derived 3D objects
- integrate image-to-3D generation into a ComfyUI pipeline
- control gaussian count to trade quality against rendering cost

## When to choose
- you need fast single-image to 3D gaussian splat generation without heavy dependencies
- you want a simple, hackable codebase to integrate into your own pipeline
- you need variable gaussian counts to balance quality and performance
- you work in ComfyUI and want an official image-to-gaussian-splat workflow

## When to avoid
- you need to train or fine-tune the model yourself (use the separate TripoSplat-Training repo)
- you need textured meshes rather than gaussian splat representations
- you require multi-view or video input rather than a single image
- you have no GPU available for inference

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, image-processing, graphics, sdk
- domain: computer-vision, graphics, artificial-intelligence
- platform: python, cross-platform
- tags: image-to-3d, gaussian-splatting, 3d-reconstruction, inference, comfyui, single-image-to-3d, game-development, gpu

## Member repositories
- VAST-AI-Research/TripoSplat (main) score 57

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:03.725378+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T08:22:20.454786+00:00, confidence not recorded.
  - readme: https://github.com/VAST-AI-Research/TripoSplat (fetched 2026-08-28T04:04:03.725378+00:00, sha 22d352a7f7e8)
- Data as of 2026-08-30T08:39:29.467469+00:00.