BiRefNet vs face-parsing

Specs, VRAM requirements and download trends — updated 24 October 2026.

ZhengPeng7 · A · Open weights
Params
220 M
Context
—
30d
751 K
mit · commercial OK
jonathandinu · B · Open weights
Params
85 M
Context
—
30d
183 K

Specification comparison

Differences are highlighted; identical values are muted.

Specification BiRefNet face-parsing
Parameters 220 M 85 M
Architecture BiRefNet SegformerForSemanticSegmentation
Context length — —
Licence mit —
Commercial use Allowed Unknown
Languages — 1
Downloads 30d 751,374 183,467
Downloads all time 20.5 M 10.6 M
Quantizations on the Hub SAFETENSORS SAFETENSORS
Gated No No
First seen on the Hub 2024-07-12 2022-07-06

Download trend

Daily snapshots, last 89 days (28 Jul – 24 Oct)

1.0 M183 K
BiRefNet face-parsing

VRAM side by side

On RTX 4090 · 24 GB · 8K context unless noted

face-parsing · fp16 0.9 GB / 24 GB
✅ Runs comfortably
BiRefNet · fp16 1.0 GB / 24 GB
✅ Runs comfortably

Adjacent comparisons