Video model · wan
WA Wan 2.1 T2V 1.3B requirements
DIT video model · 1.3B params · 832×480 (480p), 81f (~5s) · released Feb 2025. Realistic minimum to run: Nvidia GeForce RTX 3060 (12GB) at Q4 GGUF.
Backbone size by precision
| Precision | Size |
|---|---|
| fp16 / bf16 | 2.84 GB |
| Q8 GGUF | 1.54 GB |
| Q4 GGUF (recommended) | 0.98 GB |
Backbone weights only. Peak VRAM is dominated by the activation memory for 81 frames at 832×480 (480p), not the file size.
Pipeline components
| Component | Size |
|---|---|
| umT5-XXL text encoder offloaded | 3.66 GB |
| VAE (3D) | 0.51 GB |
Video VAEs are larger than image VAEs because they decode a temporal stack of frames.
Run it
Wan 2.1 T2V 1.3B runs in ComfyUI or Diffusers. Generating more frames or higher resolution raises peak VRAM sharply; the Q4 GGUF figure is for the default 81-frame clip.
Which devices can run Wan 2.1 T2V 1.3B?
Apple Silicon Macs
- Apple M1 (8GB) No
- Apple M2 (16GB) Yes
- Apple M4 (16GB) Yes
- Apple M5 (16GB) Yes
- Apple M3 Pro (18GB) Yes
- Apple M4 (24GB) Yes
- Apple M4 Pro (24GB) Yes
- Apple M5 (32GB) Yes
- Apple M4 Pro (48GB) Yes
- Apple M5 Pro (48GB) Yes
- Apple M4 Max (64GB) Yes
- Apple M4 Max (128GB) Yes
- Apple M5 Max (128GB) Yes
- Apple M3 Ultra (256GB) Yes
RAM-only laptops
No mainstream local runtime for a 1.3B video model on RAM-only laptops yet.
iPhone & iPad
No mainstream local runtime for a 1.3B video model on iPhone & iPad yet.
Android
No mainstream local runtime for a 1.3B video model on Android yet.
NVIDIA GPUs
AMD GPUs
FAQ
How much VRAM does Wan 2.1 T2V 1.3B need?
At Q4 GGUF the realistic peak is ~6 GB, versus ~20 GB with every component resident. With aggressive CPU offload it drops to ~5 GB, much slower.
Why is peak VRAM lower than the sum of the files?
The text encoder is run once to encode your prompt, then offloaded to CPU before the frames are generated, so it is not resident at the memory peak.
Can I use Wan 2.1 T2V 1.3B commercially?
Yes. Wan 2.1 T2V 1.3B is licensed Apache-2.0, which permits commercial use.
The accessible Wan tier (1.3B DiT). Backbone bf16 is 2.84GB; the umT5-XXL encoder (11.4GB) is offloaded to CPU via --t5_cpu, so the native bf16 GPU peak is ~8.2GB (Wan-AI's own figure) and GGUF Q4 + T5 on CPU drops it to ~5-6GB. Generates 832x480, 81 frames (~5s) at 16fps. Apache-2.0, commercial OK. Anchor is the GGUF Q4 community path (synthesis). Sources: Wan-AI card, samuelchristlie GGUF, city96 umT5 GGUF, Wan2.1 repo.
Sources
VRAM is a sourced peak-usage anchor at Q4 GGUF (composed from component sizes, not a single measurement) for the default clip length, validated 2026-06-15. See methodology.