Fix: disable track_tqdm (ProgressUnit crash) for API use

Persist model cache on /data (symlink) for fast cold-wakes

Native GPU load (revert offload) for full speed+quality on 48GB

Load model to CPU then enable_vram_management streams to GPU (was OOM placing on cuda)

L4/A10G fit: enable_vram_management(num_persistent_param_in_dit=0) CPU-stream DiT

Update app.py

Update app.py

Update app.py

Update app.py

Update app.py

Update app.py

Update app.py

Update app.py

Update app.py

gpu based

Update app.py

Update app.py

Update app.py

Update app.py

Update app.py

Update app.py

Update app.py

Update app.py

clean up

do faster CPU based tts

Update app.py

Update app.py

Update app.py

Update app.py

Update app.py

Update app.py

Update app.py

Update app.py

Update app.py

Update app.py

Update app.py

session based aspect ratio as oppose to the global one

Update app.py

small clean up before the new HF Space

Update app.py

unnecessary caching disabled

Update app.py

Update app.py

aspect ratio options added

cha cha cha

cha cha cha

session state fixed

more debug information

Update app.py

Update app.py