Wan 2.2 TI2V 5B in FP8 needs 11.6 GB of GPU memory or more (1280×704, 121 frames). The cheapest option right now is RTX 3060 12 GB, from $0.084/h.
“Minimum” is what ComfyUI does by default when memory is short. “No swapping” keeps everything loaded, so repeated runs are faster.
| Weights | Model | Text encoder | Minimum | No swapping | Fits on |
|---|---|---|---|---|---|
| BF16 | 9.3 | 10.6 | 16.2 | 26.8 | RTX 4000 Ada 20 GB |
| FP8 | 4.7 | 6.3 | 11.6 | 17.8 | RTX 3060 12 GB |
| GGUF Q8 | 4.9 | 6.3 | 11.8 | 18.1 | RTX A4000 16 GB |
| GGUF Q4 | 2.6 | 6.3 | 9.5 | 15.8 | RTX 3080 10 GB |
Working memory is our estimate for the listed output size and frame count. ComfyUI also runs the model on a smaller card by swapping part of the weights to system RAM, only slower.
File sizes checked against Hugging Face on 2026-10-19. 1 GB here is 1024³ bytes, like GPU memory. Hugging Face lists decimal gigabytes, so its numbers are 7% larger.
GPUs from the catalog the model fits on, by price per hour.
The price is the cheapest GPU of the model × the GPU count. A 2–8 GPU server can cost a little more per GPU.
The “Wan2GP” template starts the model in a ready environment: open the link once the server has loaded.
In ComfyUI the model goes into the models folder; the template ships with ComfyUI Manager.