Refresh card: include the 1.1 transformer and LoRA added by the delta upload

Refresh card: include the 1.1 transformer and LoRA added by the delta upload

Add distilled-1.1 transformer (int8)

Add LoRA 1.1 (mirror upstream)

Add related projects links to model card

Add usage section with ltx-2-mlx reference

Upload MLX model via mlx-forge

Fix: distilled also has cross_attention_adaln=True

Sync distilled LoRA (rank=384) from Lightricks/LTX-2.3

Rename transformer.safetensors to transformer-distilled.safetensors

Update config.json with per-variant cross_attention_adaln

Update split_model.json for multi-variant support

feat: include audio VAE encoder weights (decoder + encoder + stats)

fix: transpose vocoder STFT basis buffers (forward_basis, inverse_basis) to MLX channels-last

fix: transpose vocoder buffers (filter, basis) and upscaler kernel to MLX channels-last layout

Remove unversioned upscaler file (replaced by versioned name)

Remove unversioned upscaler file (replaced by versioned name)

Remove unversioned upscaler file (replaced by versioned name)

Remove unversioned upscaler file (replaced by versioned name)

Remove unversioned upscaler file (replaced by versioned name)

Remove unversioned upscaler file (replaced by versioned name)

Rename upscaler files with version (x2_v1_1, x1_5_v1_0, x2_v1_0)

Add spatial upscalers (x2, x1.5) and temporal upscaler (x2) for multi-stage pipelines

fix: pre-convert ConvTranspose1d (ups) weights to MLX layout (O,K,I)

Upload MLX model via mlx-forge

initial commit