nodes/1038lab-ComfyUI-FlashVSR
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Update update.md

Browse Source
This commit is contained in:
AI Lab
2025-11-17 15:03:38 -08:00
committed by GitHub
parent 2fd1afb910
commit f1adf640e1
1 changed files with 6 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
update.md
+6
View File
@@ -1,5 +1,10 @@
# ComfyUI-FlashVSR Update Log
## V1.1.1 (2025/11/17)
- Refresh FlashVSR 1.1 main model package (renamed to `FlashVSR1_1.safetensors` alongside `LQ_proj_in.safetensors` and `TCDecoder.safetensors`) and update latest HuggingFace models.
- Fix the core loader to validate the new filenames and raise clear warnings when any required asset is missing, preventing silent fallbacks.
- Fix runtime RoPE dtype patch so FlashVSR runs on Apple Silicon/MPS without float64 errors.
## V1.1.0 (2025/11/15)
### 🔧 Fixes
@@ -15,3 +20,4 @@
This model improves T2V → VSR sharpness, detail preservation, and temporal stability.