are you able to share tests ?
#2
by skylava - opened
Please share test samples or scores if possible or at least add a brief to the docs , much appreciated !
[INFO] gguf qtypes: F32 (269), F16 (16), Q4_K (250)
[INFO] model weight dtype torch.bfloat16, manual cast: None
[INFO] model_type FLOW
[INFO] Requested to load MiniMaxH3
[INFO] loaded completely; 19440.97 MB usable, 19440.07 MB loaded, full load: True
100%|██████████████████████████████████████████████████████████████████████████| 25/25 [07:45<00:00, 18.63s/it]
[INFO] Requested to load MiniMaxH3AudioVAE
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.
[INFO] Requested to load MiniMaxH3VideoVAE
[INFO] Unloaded partially: 3987.76 MB freed, 15452.31 MB remains loaded, 186.05 MB buffer reserved, lowvram patches: 0
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.
[INFO] Prompt executed in 584.42 seconds
Used the Q4KM model and its Q4KM text encoder version to get things running.