# 3DTopia/LGM

[ECCV 2024 Oral] LGM: Large Multi-View Gaussian Model for High-Resolution 3D Content Creation.

Repository: https://github.com/3DTopia/LGM
Canonical: https://ross.abutalabs.com/products/lgm
Homepage: https://me.kiui.moe/lgm/
Language: Python
License: MIT
License Family: permissive
Topics: gaussian-splatting, image-to-3d, text-to-3d
Last push: 2024-08-20T16:27:45+00:00

## Health v2 (maintenance only)
Score: 26/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 67
- inputs: {"age_days": 939, "days_push": 743, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2111, forks 140 (observed 2026-08-28T04:06:14.265600+00:00)

## What it is
LGM is the official PyTorch implementation of an ECCV 2024 Oral paper that generates high-resolution 3D models from text prompts or single-view images using a Large Multi-view Gaussian Model. It fuses multi-view Gaussian features produced by diffusion models and renders them differentiably, exporting 3D objects in about 5 seconds.

## Use cases
- generate a 3D model from a text prompt
- convert a single image into a 3D asset
- create gaussian splat 3D objects in seconds
- export generated 3D content as meshes for games
- run a gradio demo for text/image to 3D
- visualize generated 3D gaussian ply files in a local gui

## When to choose
- you need fast feed-forward text-to-3D or image-to-3D generation
- you want high-resolution 3D output via gaussian splatting
- you need a research baseline for multi-view gaussian 3D generation
- you want pretrained weights and ready-made demos (HuggingFace, Replicate)

## When to avoid
- you need production-grade, artist-ready topology meshes without cleanup
- you lack a GPU with ~10GB memory for full inference
- you need precise CAD-like or parametric 3D modeling
- you want a maintained tool with long-term support rather than a research codebase

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, image-processing, graphics, simulation
- domain: artificial-intelligence, deep-learning, computer-vision, graphics
- platform: python, cross-platform
- tags: gaussian-splatting, image-to-3d, text-to-3d, 3d-generation, multi-view-diffusion, eccv-2024, 3d-content-creation, game-development, gpu, linux

## Member repositories
- 3DTopia/LGM (main) score 26

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:14.265600+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 2026-08-30T02:53:42.393892+00:00, confidence not recorded.
  - readme: https://github.com/3DTopia/LGM (fetched 2026-08-28T04:06:14.265600+00:00, sha 49e8820ac757)
  - homepage: https://me.kiui.moe/lgm/ (fetched 2026-08-29T10:33:43.069759+00:00, sha e2e3a52b0e62)
- Data as of 2026-08-30T08:39:29.467469+00:00.