My 4steps workflow without promt. I use it for scaling and generating frames in video on my rx 6800xt graphics card.Scaling and generating frames for a 33-frame video takes between ~160 and ~480 seconds depending on resolution after the models are fully loaded. The models are small so it doesn't take long.It's not as fast, but it provides good quality.All required models download:<a target="_blank" rel="ugc" href="https://civarchive.com/models/1295569?modelVersionId=1462182">Kijai-Wan2.1-T2V-1.3B</a><a target="_blank" rel="ugc" href="https://huggingface.co/Kijai/WanVideo_comfy/blob/main/Wan21_CausVid_bidirect2_T2V_1_3B_lora_rank32.safetensors">CausVid 13B lora</a><a target="_blank" rel="ugc" href="https://huggingface.co/chatpig/encoder/blob/main/umt5_xxl_fp8_e4m3fn_scaled.safetensors">umt5_xxl_fp8_e4m3fn_scaled</a><a target="_blank" rel="ugc" href="https://huggingface.co/Comfy-Org/Wan_2.1_ComfyUI_repackaged/blob/main/split_files/vae/wan_2.1_vae.safetensors">Wan_2.1_vae</a>For v2.0 download any upscale model you like from here <a target="_blank" rel="ugc" href="https://huggingface.co/LS110824/upscale/tree/main">upscalers</a>Required custom nodes:ComfyUI-KJnodesVideoHelperSuiteComfiUI-Frame-Interpolation

Wan2.1 Upscaler + FrameGen

Added image upscale using model instead of lanzcos.Special thx to AI Artistry Atelier. I borrowed an idea from his video on youtube.

Files with SHA256: 79501861acc10f3785b3ba32c1933af3c50bf3c3c70f339081c5921c97a1f092