SkyReels-V1 Img2Vid Simple Workflow (Finetuned Hunyuan Video)

Version 3 (Q4)

Update ComfyUI and all nodes to the latest version.

If your VRAM is 16GB or less, try reducing the number of frames.

If the results are not good, you can try the following:

1. Change the control_after_generate of the Pass 1 node to increment or randomize.

2. Modify the prompt or use a different prompt

3. Adding "FPS-24, " to the prompt

4. Increase the resolution a little

5. Increase the steps

List used:

comfyui-logic

rgthree-comfy

ComfyUI-GGUF

ComfyUI-KJNodes

comfyui-custom-scripts

ComfyUI-VideoHelperSuite

comfyui-frame-interpolation

ComfyUI-HunyuanVideoMultiLora

MODEL: https://huggingface.co/Kijai/SkyReels-V1-Hunyuan_comfy/blob/main/skyreels-hunyuan-I2V-Q4_K_S.gguf

Put models/diffusion_models/

VAE: https://huggingface.co/Kijai/HunyuanVideo_comfy/blob/main/hunyuan_video_vae_bf16.safetensors

Put models/vae/

CLIP 1: https://huggingface.co/Comfy-Org/HunyuanVideo_repackaged/blob/main/split_files/text_encoders/clip_l.safetensors

Put models/clip/

CLIP 2: https://huggingface.co/Comfy-Org/HunyuanVideo_repackaged/blob/main/split_files/text_encoders/llava_llama3_fp8_scaled.safetensors

Put models/clip/

Version 2 (Q5)

ComfyUI-GGUF

comfyui-logic

ComfyUI-KJNodes

ComfyUI-VideoHelperSuite

comfyui-florence2

comfyui-custom-scripts

comfyui-frame-interpolation

MODEL: https://huggingface.co/Kijai/SkyReels-V1-Hunyuan_comfy/blob/main/skyreels-hunyuan-I2V-Q5_K_M.gguf

Put models/diffusion_models/

VAE: https://huggingface.co/Kijai/HunyuanVideo_comfy/blob/main/hunyuan_video_vae_bf16.safetensors

Put models/vae/

CLIP 1: https://huggingface.co/Comfy-Org/HunyuanVideo_repackaged/blob/main/split_files/text_encoders/clip_l.safetensors

Put models/clip/

CLIP 2: https://huggingface.co/Comfy-Org/HunyuanVideo_repackaged/blob/main/split_files/text_encoders/llava_llama3_fp8_scaled.safetensors

Put models/clip/

Version 1

32GB RAM / 24GB VRAM

ComfyUI-KJNodes

ComfyUI-VideoHelperSuite

comfyui-florence2

comfyui-custom-scripts

comfyui-frame-interpolation

MODEL: https://huggingface.co/Kijai/SkyReels-V1-Hunyuan_comfy/blob/main/skyreels_hunyuan_i2v_bf16.safetensors

Put models/diffusion_models/

VAE: https://huggingface.co/Kijai/HunyuanVideo_comfy/blob/main/hunyuan_video_vae_bf16.safetensors

Put models/vae/

CLIP 1: https://huggingface.co/Comfy-Org/HunyuanVideo_repackaged/blob/main/split_files/text_encoders/clip_l.safetensors

Put models/clip/

CLIP 2: https://huggingface.co/Comfy-Org/HunyuanVideo_repackaged/blob/main/split_files/text_encoders/llava_llama3_fp16.safetensors

Put models/clip/