clipseg-rd64-refined vs mask2former-swin-large-cityscapes-semantic
Specs, VRAM requirements and download trends — updated 6 November 2026.
CIDAS · A · Open weights
Params
150 M
Context
—
30d
437 K
apache-2.0 · commercial OK
facebook · B · Open weights
Params
220 M
Context
—
30d
230 K
Specification comparison
Differences are highlighted; identical values are muted.
| Specification | clipseg-rd64-refined | mask2former-swin-large-cityscapes-semantic |
|---|---|---|
| Parameters | 150 M | 220 M |
| Architecture | CLIPSegForImageSegmentation | Mask2FormerForUniversalSegmentation |
| Context length | — | — |
| Licence | apache-2.0 | other |
| Commercial use | Allowed | Unknown |
| Languages | — | — |
| Downloads 30d | 436,988 | 229,840 |
| Downloads all time | 153.2 M | 11.1 M |
| Quantizations on the Hub | SAFETENSORS | SAFETENSORS |
| Gated | No | No |
| First seen on the Hub | 2022-11-01 | 2023-01-05 |
Download trend
Daily snapshots, last 90 days (13 Aug – 10 Nov)
1.3 M230 K
clipseg-rd64-refined
mask2former-swin-large-cityscapes-semantic
VRAM side by side
On RTX 4090 · 24 GB · 8K context unless noted
clipseg-rd64-refined · fp16
1.2 GB / 24 GB
mask2former-swin-large-cityscapes-semantic · fp16
1.5 GB / 24 GB