WAN needs about 16.2 GB of memory at Q6_K (a 12.5 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 6 quants, from 4.7 GB (Q6_K) to 18.7 GB (Q8_K).
WAN is a 14.3B-parameter Wan model in GGUF format, quantized by digitalyouneed. Its trained context length is 512 tokens. The repository has 6 quant files, from 4.7 GB (Q6_K) to 18.7 GB (Q8_K).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q6_K | 4.7 GB | 5.4 GB | Fits | Get file → | |
| Q8_K | 6.0 GB | 6.7 GB | Fits | Get file → | |
| Q6_K | 12.5 GB | 13.2 GB | Too big | Get file → | |
| Q6_K | 14.5 GB | 15.2 GB | Too big | Get file → | |
| Q8_K | 15.9 GB | 16.6 GB | Too big | Get file → | |
| Q8_K | 18.7 GB | 19.4 GB | Too big | Get file → |
What GPU do you need for WAN?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: no quant fits at an 8K context.
- 6 GB VRAM: no quant fits at an 8K context.
- 8 GB VRAM: no quant fits at an 8K context.
- 12 GB VRAM: the best quant that fits is Q8_K (6.0 GB file, about 9.8 GB in use).
- 16 GB VRAM: the best quant that fits is Q8_K (6.0 GB file, about 9.8 GB in use).
- 24 GB VRAM: the best quant that fits is Q8_K (18.7 GB file, about 22.4 GB in use).
Which WAN quant should you choose?
Q6_K (12.5 GB) is the middle option in this repository. Very close to the original quality for most uses, at a clearly larger size than 4-bit.
If memory is tight, the smallest file is Q6_K at 4.7 GB, with the quality trade-offs that come with fewer bits. Very close to the original quality for most uses, at a clearly larger size than 4-bit.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about WAN
How much VRAM does WAN need?
About 16.2 GB at Q6_K with an 8K context window. That is the 12.5 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run WAN on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 8.4 GB, so you would need a larger GPU or CPU offloading.
What is the file size of WAN GGUF?
It depends on the quant: Q6_K is 4.7 GB, Q8_K is 6.0 GB, Q6_K is 12.5 GB, Q6_K is 14.5 GB, Q8_K is 15.9 GB, Q8_K is 18.7 GB.
What context length does WAN support?
The GGUF header lists a trained context length of 512 tokens (512). The KV cache grows with the context you use: about 3.2 GB at 8K and 0.2 GB at 512.
Where can I download WAN GGUF?
The files are on Hugging Face in the digitalyouneed/WAN repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.