fashn-AI/fashn-vton-1.5

FASHN VTON v1.5: Efficient Maskless Virtual Try-On in Pixel Space

52
/ 100
Established

Leverages DWPose for human pose detection and a human parser for garment localization, eliminating the need for manual segmentation masks. The model operates directly in pixel space using bfloat16 precision on modern GPUs (Ampere+) while gracefully falling back to float32 on older hardware. Supports three garment categories (tops, bottoms, one-pieces) and accepts both model photos and flat-lay product images as input.

166 stars.

No Package No Dependents
Maintenance 10 / 25
Adoption 10 / 25
Maturity 11 / 25
Community 21 / 25

How are scores calculated?

Stars

166

Forks

33

Language

Python

License

Last pushed

Feb 01, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/generative-ai/fashn-AI/fashn-vton-1.5"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.