Model card: 16 GB / 22 GB GPUs (ComfyUI workflows at a 14 GiB cap, inference.py --low-vram)

README.md

@@ -62,7 +62,7 @@ token-refiner blocks (312 modules). alpha = rank = 128. The safetensors metadata
 
 The inference code lives in the [code repository](https://github.com/Yzmblog/DMAD): `inference.py` with the sampler
 the paper used (re-noise step rule) and a Diffusers-pipeline example; its README covers the environment. Sampling
-settings: 4 steps, time shift 12 (video) and 2 (audio), no classifier-free guidance, 124 frames at 24 fps.
+settings: 4 steps, time shift 12 (video) and 2 (audio), no classifier-free guidance, 124 frames at 24 fps. `--low-vram` runs it on a 16 GB GPU.
 
 ```bash
 git clone https://github.com/Yzmblog/DMAD.git && cd DMAD   # code + environment setup (see its README)
@@ -120,6 +120,13 @@ ComfyUI-Manager), which both workflows include. Without it, 5 s (length 124) fit
 the one made with the memory node. Times are end to end (model loading, text encoding, sampling, decoding) on an H200
 with the PyTorch allocator capped at 24 GiB; a consumer GPU is slower, and sampling time scales linearly with steps.
 
+**16 GB and 22 GB GPUs.** Both workflows also run unchanged with the GPU memory capped at 14 GiB (room for a 16 GB
+card's desktop and CUDA context): ComfyUI's dynamic VRAM keeps less of the model resident and streams more of it from
+system memory, and the videos are bit-identical to the 24 GiB runs. About 46 GiB of system memory is in use, so 64 GB
+of RAM is recommended. Outside ComfyUI, `inference.py --low-vram` in the
+[code repository](https://github.com/Yzmblog/DMAD) makes 5 s and 15 s videos in under 14 GiB of GPU memory, bit-identical
+to its larger-GPU runs.
+
 ## Wan2.1 (text-to-video)
 
 Full generators (EMA, spectral norm folded in) in the `.pth` format of