RMBG-2.0 vs mask2former-swin-large-ade-semantic
Specs, VRAM requirements and download trends — updated 10 November 2026.
Specification comparison
Differences are highlighted; identical values are muted.
| Specification | RMBG-2.0 | mask2former-swin-large-ade-semantic |
|---|---|---|
| Parameters | 221 M | 220 M |
| Architecture | BiRefNet | Mask2FormerForUniversalSegmentation |
| Context length | — | — |
| Licence | other | other |
| Commercial use | Unknown | Unknown |
| Languages | — | — |
| Downloads 30d | 407,616 | 335,778 |
| Downloads all time | 11.0 M | 5.3 M |
| Quantizations on the Hub | SAFETENSORS | SAFETENSORS |
| Gated | Yes — accept licence | No |
| First seen on the Hub | 2024-10-29 | 2023-01-05 |
Download trend
Daily snapshots, last 90 days (13 Aug – 10 Nov)
694 K273 K
RMBG-2.0
mask2former-swin-large-ade-semantic
A overtook B on 9 Nov
VRAM side by side
On RTX 4090 · 24 GB · 8K context unless noted
RMBG-2.0 · fp16
1.5 GB / 24 GB
mask2former-swin-large-ade-semantic · fp16
1.5 GB / 24 GB