Skip to content
localmodel.run

video model · wan · Linux

Can I run Wan 2.1 T2V 1.3B on Nvidia DGX Spark (128GB)?

Compatibility verdict VRAM check
Yes, it runs fast on this GPU

Yes. Wan 2.1 T2V 1.3B runs on Nvidia DGX Spark (128GB) at Q4 GGUF (~6 GB of ~112 GB usable).

Needs ~6 GB Device usable ~112 GB

Runs at Q4 GGUF using ~6 GB of ~112 GB usable.

Peak VRAM
~6 GB
Usable on device
~112 GB
Device memory
128 GB
Quant
Q4 GGUF
Share on X Share on Reddit

How to run it

Use ComfyUI or Diffusers at Q4 GGUF. 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.

Model wan
Type
video (DIT)
Parameters
1.3B
Peak VRAM
~6 GB at Q4 GGUF
Resolution
832×480 (480p)
License
Apache-2.0
Full Wan 2.1 T2V 1.3B requirements →
Device Linux
Memory
128 GB unified
Usable for weights
~112 GB
Power draw
~140 W
Best runtime
llama.cpp (CUDA) / Ollama (CUDA)
Best models for Nvidia DGX Spark (128GB) →

You could also run

Run Wan 2.1 T2V 1.3B on other hardware

Nvidia GeForce RTX 3060 Ti (8GB)Nvidia GeForce GTX 1070 (8GB)Nvidia GeForce RTX 3060 (12GB)Nvidia GeForce RTX 4070 (12GB)Nvidia GeForce RTX 5070 (12GB)Apple M2 (16GB)

FAQ

Can Nvidia DGX Spark (128GB) run Wan 2.1 T2V 1.3B?

Yes. Wan 2.1 T2V 1.3B runs on Nvidia DGX Spark (128GB) at Q4 GGUF (~6 GB of ~112 GB usable).

How much VRAM does Wan 2.1 T2V 1.3B need?

Nvidia DGX Spark (128GB) has room to spare. At Q4 GGUF the realistic peak is ~6 GB of VRAM, versus ~20 GB with every component kept resident (no offload). With aggressive CPU offload it drops to ~5 GB, much slower.

What do I use to run Wan 2.1 T2V 1.3B locally?

Wan 2.1 T2V 1.3B 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; catalog updated 2026-11-01. See methodology.