GGUF is slow
Open
- Dominant language
- Python
- Stars
- 133k
- Forks
- 15.7k
- Avg merge
- 1d 7h
- Merged PRs (30d)
- 158
Description
@rattus128
rtx 4060 8GB 32GB ram windows 10, last comfyui git
ltx2.3 gguf Q5
[INFO] Unloaded partially: 1233.75 MB freed, 2901.02 MB remains loaded, 308.22 MB buffer reserved, lowvram patches: 85
38%|███████████████████████████████▌ | 3/8 [04:48<07:03, 84.72s/it]
in kj memory display LTXAV is STATIC.
same with ltx2.3 nvfp4
[INFO] Model LTXAV prepared for dynamic VRAM loading. 17833MB Staged. 1344 patches attached. Force pre-loaded 608 weights: 3303 KB.
100%|████████████████████████████████████████████████████████████████████████████████████| 8/8 [04:08<00:00, 31.07s/it]
seems GGUF dont use dynamic vram, and is very slow
Why upload image here fails?
Contributor guide
Assessment
This issue has not been assessed yet.