Downloads · 30 days
0
LTX-2.3-distilled-1.1-MLX-8bit
Ghost13lade/LTX-2.3-distilled-1.1-MLX-8bit
LTX-2.3-distilled-1.1-MLX-8bit is a machine learning model from Ghost13lade. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
MLX-converted and 8-bit transformer-quantized checkpoint for LTX-2.3 distilled v1.1.
Access
Public
Updated Jul 14, 2026
Repo size
37.3 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors37.3 GB · 100%
At a glance
- Access
- Public
- Created
- Jul 14, 2026
- Updated
- Jul 14, 2026
- SHA
- 060085bf
Model card
From the Hugging Face model README
LTX-2.3 Distilled v1.1 MLX 8-bit
MLX-converted and 8-bit transformer-quantized checkpoint for LTX-2.3 distilled v1.1.
For use with mlx-video on Apple Silicon.
Source model: Lightricks/LTX-2.3 Source checkpoint: ltx-2.3-22b-distilled-1.1.safetensors Included upscaler: ltx-2.3-spatial-upscaler-x2-1.1.safetensors
Quantization: transformer 8-bit, group_size=64. VAE/audio/vocoder/text projections are converted MLX weights.
Example:
python -m mlx_video.models.ltx_2.generate --model-repo Ghost13lade/LTX-2.3-distilled-1.1-MLX-8bit --prompt "A calm ocean wave at sunset, cinematic" --height 512 --width 768 --num-frames 33
Tested locally with mlx-video: T2V 768x512, 33 frames, around 31GB peak memory.
- Created
- Jul 14, 2026
- Updated
- Jul 14, 2026