Wan 2.2 Workflow for 4GB VRAM cards, tested in a GTX 1050 TiHello beautiful people.Disclaimer: You need the patience and persistence of a fisherman to run this, if you have and old GPU.This is according to my experience, the only workflow i2v that works with my 1050 ti 4GB VramRecommended resolutions 512x512 and 600x400 or 400x600.I know there are several T2v 5b models out there but some nodes used in workflows don´t work with old nvidia cards.What can you expect?At first run will be a cuda OOM(*), just click in run button again and should be fine.At 16 FPS49 frames jobs take 50 min. to finish81 frames jobs take 1 hour 45 min. to finish145 frames about 2 hours 40 min. to finishAll times are aproximations, your mileage may vary.Why I'm share this?To encourage people and curious minds to use the wan2.2 model and give it a try. I hope this workflow helps someone.Hardware testedIntel(R) Core(TM) i5-9400F CPU @ 2.90GHz, 16 GB Ram, nvidia card Gtx 1050 Ti 4GB Vram, 1TB SSD disk, virtual memory 64GB in fixed size Models used in WorkflowModel Q4_K_M Rapid V9: <a target="_blank" rel="ugc" href="https://civarchive.com/models/1855105?modelVersionId=2137014">https://civarchive.com/models/1855105?modelVersionId=2137014</a>wan_2.1_vae.safetensors:<a target="_blank" rel="ugc" href="https://huggingface.co/Comfy-Org/Wan_2.1_ComfyUI_repackaged/tree/main/split_files/vae">https://huggingface.co/Comfy-Org/Wan_2.1_ComfyUI_repackaged/tree/main/split_files/vae</a>umt5-xxl-encoder-Q8_0.gguf:<a target="_blank" rel="ugc" href="https://huggingface.co/city96/umt5-xxl-encoder-gguf/tree/main">https://huggingface.co/city96/umt5-xxl-encoder-gguf/tree/main</a>(*)PD: if you have another OOM just press that run button again.

Files with SHA256: 3b690fe9ebb60ef10f436c782da9ce286b773b7cfb3ebe31e5bb15dbbead58df