video model · ltx-video · macOS
LX Can I run LTX-Video 13B on Apple M4 Pro (48GB)?
Yes. LTX-Video 13B runs on Apple M4 Pro (48GB) at fp8 (~20 GB of ~32 GB usable).
Runs at fp8 using ~20 GB of ~32 GB usable.
- Peak VRAM
- ~20 GB
- Usable on device
- ~32 GB
- Device memory
- 48 GB
- Quant
- fp8
How to run it
Use ComfyUI or Diffusers at fp8. The big text encoder is loaded to encode your prompt, then offloaded before generation, which is why peak VRAM stays near the backbone size rather than the sum of every file.
- Type
- video (DIT)
- Parameters
- 13B
- Peak VRAM
- ~20 GB at fp8
- Resolution
- 1216×704
- License
- LTX-Video Open Weights (OpenRAIL-M)
- Memory
- 48 GB unified
- Usable for weights
- ~32 GB
- Best runtime
- Ollama (MLX backend) / MLX direct
You could also run
Run LTX-Video 13B on other hardware
FAQ
Can Apple M4 Pro (48GB) run LTX-Video 13B?
Yes. LTX-Video 13B runs on Apple M4 Pro (48GB) at fp8 (~20 GB of ~32 GB usable).
How much VRAM does LTX-Video 13B need?
Apple M4 Pro (48GB) has room to spare. At fp8 the realistic peak is ~20 GB of VRAM, versus ~38 GB with every component kept resident (no offload). With aggressive CPU offload it drops to ~12 GB, much slower.
What do I use to run LTX-Video 13B locally?
LTX-Video 13B runs in ComfyUI or Diffusers. It loads as a video diffusion checkpoint plus its text encoder and VAE, not a single chat command.
Sources
VRAM figures are sourced peak-usage anchors at the noted quant, validated 2026-06-15. See methodology.