TencentARC/InstantMesh

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

4509 stars · 501 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

InstantMesh is a feed-forward framework for generating 3D meshes from a single image using sparse-view large reconstruction models (LRM/Instant3D architecture). It provides inference and training code, model weights, and a Gradio demo.

Use cases

• generate a 3d mesh from a single photo
• image to 3d model conversion
• create 3d assets for games from images
• reconstruct 3d geometry from one view
• fine-tune multi-view diffusion models for 3d generation
• run a local image-to-3d demo

When to choose

• you need fast feed-forward single-image to 3D mesh generation
• you want pretrained model weights and a ready Gradio demo
• you want to fine-tune Zero123++ or train sparse-view reconstruction models

When to avoid

• you need high-fidelity textured 3D scans from real multi-view captures
• you have no GPU available
• you need parametric CAD modeling rather than mesh generation

Facets

library · maturity active

machine-learning deep-learning image-processing graphics llm-inference artificial-intelligence computer-vision graphics machine-learning deep-learning python 3d-reconstruction image-to-3d mesh-generation lrm multi-view-diffusion gradio-demo pytorch gpu linux docker

Member repositories

For agents

markdown · JSON · MCP: product_card(name="TencentARC/InstantMesh")