BiRefNet vs mask2former-swin-large-ade-semantic

Specs, VRAM requirements and download trends — updated 6 August 2026.

ZhengPeng7 · A
Params
220 M
Context
30d
715 K
mit · commercial OK
facebook · B
Params
220 M
Context
30d
625 K

Specification comparison

Differences are highlighted; identical values are muted.

Specification BiRefNet mask2former-swin-large-ade-semantic
Parameters 220 M 220 M
Architecture BiRefNet Mask2FormerForUniversalSegmentation
Context length
Licence mit other
Commercial use Allowed Unknown
Languages
Downloads 30d 715,119 624,580
Downloads all time 18.1 M 4.1 M
Quantizations on the Hub SAFETENSORS SAFETENSORS
Gated No No
First seen on the Hub 2024-07-12 2023-01-05

Download trend

Daily snapshots, last 10 days (28 Jul – 6 Aug)

765 K621 K
BiRefNet mask2former-swin-large-ade-semantic

VRAM side by side

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

BiRefNet · fp16 1.0 GB / 24 GB
✅ Runs comfortably
mask2former-swin-large-ade-semantic · fp16 1.5 GB / 24 GB
✅ Runs comfortably

Adjacent comparisons