Subodh92290/latentsync-h100
Upload folder using huggingface_hub
Upload folder using huggingface_hub
Optimize chunk size to 4s for ultra-fast and reliable ZeroGPU diffusion
Set duration=45 on ZeroGPU inference
Run LatentSync in-process for direct ZeroGPU CUDA execution
Add automatic ZeroGPU chunking pipeline for long videos
Add ffmpeg-python and LatentSync dependencies to requirements.txt
Set duration=60 on ZeroGPU inference
Refactor spaces.GPU decorator to GPU-only inference step with duration=120
Fix launch keyword args and ensure full LatentSync codebase
Fix LatentSync repo codebase initialization and inference paths
Upload app.py with huggingface_hub
Upload app.py with huggingface_hub
Upload app.py with huggingface_hub
Upload app.py with huggingface_hub
Fix boolean schema bug and enable API queue endpoint
Deploy LatentSync 1.6 HD Turbo ZeroGPU H100
Deploy LatentSync 1.6 HD Turbo ZeroGPU H100
Deploy LatentSync 1.6 HD Turbo ZeroGPU H100
Deploy LatentSync 1.6 HD Turbo ZeroGPU H100
Deploy LatentSync 1.6 HD Turbo ZeroGPU H100
Deploy LatentSync 1.6 HD Turbo ZeroGPU H100
Deploy LatentSync 1.6 HD Turbo ZeroGPU H100
Deploy LatentSync 1.6 HD Turbo ZeroGPU H100
Deploy LatentSync 1.6 HD Turbo ZeroGPU H100
Deploy LatentSync 1.6 HD Turbo ZeroGPU H100
Deploy LatentSync 1.6 HD Turbo ZeroGPU H100
initial commit
