Model reference · open weights
SANA-WM_chunk_causal is an open-weight video model from Efficient-Large-Model. SANA-WM_chunk_causal (BF16) weighs 10.7 GB; the smallest configuration that runs it is RTX 3060 12 GB.
Summary of the Efficient-Large-Model/SANA-WM_chunk_causal model card, 2026-10-01
What it is
| Released by | Efficient-Large-Model |
|---|---|
| Released | 2026-06-10 |
| VRAM | 10.7 GB for the weights |
What it runs on
| Card | Weights | Memory |
|---|---|---|
| RTX 3060 12 GB | tight | 11.6 GB |
| RTX 4060 Ti 16 GB | fits | 15.4 GB |
| RTX 3090 24 GB | fits | 23.4 GB |
| RTX 4090 24 GB | fits | 23.4 GB |
| RTX 5090 32 GB | fits | 31.0 GB |
| L40S 48 GB | fits | 44.0 GB |
| A100 80 GB | fits | 78.2 GB |
| H100 80 GB | fits | 78.1 GB |
| RTX PRO 6000 Blackwell 96 GB | fits | 93.8 GB |
| DGX Spark (GB10) 128 GB unified | fits | 107 GB |
| H200 141 GB | fits | 138 GB |
| B200 180 GB | fits | 176 GB |
From the model card
This repository contains the chunk-causal Stage-1 teacher DiT checkpoint for SANA-WM.
It is intended to be used with the public SANA-WM code and reuses the VAE and refiner from Efficient-Large-Model/SANA-WM_bidirectional.
config.yaml: inference config with scheduler.vis_sampler: chunk_flow_euler.dit/sana_wm_chunk_causal_1600m_720p.safetensors: chunk-causal Stage-1 teacher checkpoint.python inference_video_scripts/wm/inference_sana_wm.py \
--config hf://Efficient-Large-Model/SANA-WM_chunk_causal/config.yaml \
--model_path hf://Efficient-Large-Model/SANA-WM_chunk_causal/dit/sana_wm_chunk_causal_1600m_720p.safetensors \
--image asset/sana_wm/demo_0.png \
--prompt asset/sana_wm/demo_0.txt \
--action "w-80,jw-40,w-40,lw-60,w-100" \
--num_frames 961 \
--output_dir results/sana_wm_chunk_causal
The sampler uses interval_k = 1 / num_chunks by default.
Quoted from the model card on Hugging Face. The full card is behind the Hugging Face link above.
How it works
Running it yourself
Rent a machine by the hour. ComfyUI is installed on it. Open ComfyUI through the tunnel and load the workflow from the model's card on Hugging Face; choose this model's file in its loader.
# on your rented machine (the ssh line is on its page in the console)
# get REPO FILE FOLDER: one file into /workspace/models/FOLDER, where ComfyUI loads it from
get() { hf download "$1" "$2" --local-dir /workspace/hf-files && mkdir -p "/workspace/models/$3" && mv "/workspace/hf-files/$2" "/workspace/models/$3/$4"; }
# the model (9.9 GB)
get Efficient-Large-Model/SANA-WM_chunk_causal dit/sana_wm_chunk_causal_1600m_720p.safetensors diffusion_models
start-comfyui# on your computer, in a second terminal: ComfyUI in your browser at http://localhost:8188
# HOST and PORT are your machine's, from its page in the console
ssh -L 8188:localhost:8188 dev@HOST -p PORT