Not sure what is wrong with this #19
Description
uv run mlx_video.generate --prompt "Two dogs of the poodle breed wearing sunglasses, close up, cinematic, sunset" -n 100 --width 768
🎬 [T2V] Generating 768x512 video with 97 frames
Prompt: Two dogs of the poodle breed wearing sunglasses, close up, cinematic, sunset
📝 Loading text encoder...
Text encoder loaded successfully
🤖 Loading transformer...
⚡ Stage 1: Generating at 384x256 (8 steps)...
🔍 Upsampling latents 2x...
Loading spatial upsampler from /Users/testing/.cache/huggingface/hub/models--Lightricks--LTX-2/snapshots/47da56e2ad66ce4125a9922b4a8826bf407f9d0a/ltx-2-spatial-upscaler-x2-1.0.safetensors...
Detected mid_channels: 1024
Loaded 73 weights
Loading VAE decoder from /Users/testing/.cache/huggingface/hub/models--Lightricks--LTX-2/snapshots/47da56e2ad66ce4125a9922b4a8826bf407f9d0a/ltx-2-19b-distilled.safetensors...
Auto-detected timestep_conditioning=False from weights
Loaded latent mean: shape (128,)
Loaded latent std: shape (128,)
Found 90 decoder weights
Found 0 timestep conditioning weights
VAE decoder loaded successfully
⚡ Stage 2: Refining at 768x512 (3 steps)...
🎞️ Decoding video...
Tiling (auto): spatial=384px, temporal=64f
/Users/testing/.local/share/uv/python/cpython-3.12.11-macos-aarch64-none/lib/python3.12/multiprocessing/resource_tracker.py:279: UserWarning: resource_tracker: There appear to be 1 leaked semaphore objects to clean up at shutdown
warnings.warn('resource_tracker: There appear to be %d '