Pin torch==2.9.1 and torchaudio==2.9.1 to fix ABI mismatch

Update default prompt to generic animation prompt

Update default prompt to Parisian café scene

Clean up public duration presets to 6s, 8s, 10s, 12s only

Revert "Standardize duration presets to 5s increments up to 60s and extend get_duration tiers"

Standardize duration presets to 5s increments up to 60s and extend get_duration tiers

Fix AudioProcessor init: use target_sample_rate kwarg and Audio object for waveform_to_mel

Add torchcodec dependency for torchaudio.load() audio backend

Default enhance_prompt back to true

Add generation metadata card with seed, resolution, timing, and enhanced prompt display

Default enhance_prompt to false

Show enhance prompt checkbox and seed/randomize controls in UI

Fix AudioDropUpload: use raw string for js_on_load to prevent double-brace JS syntax error

Rename Space title to LTX-2.3 Video [Turbo]

Enable loop on output video player

Extend duration presets to 12s with scaled GPU lease times

Default to 5s duration and 9:16 aspect ratio

Update default prompt and add alexnasa UI credit in footer

Upgrade to Gradio 6.2.0 for html_template/js_on_load support in custom components

Full UI rewrite: fancy alexnasa-style components, audio input, interpolation, social links

Add Gemma license notice to UI footer and README duplication instructions

Polish README for public release: add tags, model link, features section, performance stats

Fix get_duration arg count mismatch; add show_error=True to launch

Vendor ltx-core/ltx-pipelines and rewrite app.py for ZeroGPU compatibility

fix: use total_memory instead of total_mem for CUDA device properties

Initial turbo Space: ZeroGPU-compatible LTX-2.3 with FP8 quantization and two-phase GPU leasing

initial commit