Modified workflow from the AI ​​Verse model: <a rel="ugc" href="https://www.aiverseblog.site/2025/01/fast-hunyuan-video-lora.html">Click here to see the AI ​​Verse version</a>.<h3 id="i-tested-this-workflow-on-an-rtx-3060-video-card-with-12gb-and-34gb-of-ram-at-480x848-and-576x1024-resolutions.-ccdv0hqfn">I tested this workflow on an RTX 3060 video card with 12GB and 34GB of RAM, at 480x848 and 576x1024 resolutions.</h3>I modified the model loader and started using the Load Diffusion Model. I installed the Power Lora Loader (rgthree). For export, I started using Video Combine VHS.It works, but it is recommended not to have other activities in the background, as it required all the system resources.Base model: <a rel="ugc" href="https://huggingface.co/Kijai/HunyuanVideo_comfy/blob/main/hunyuan_video_FastVideo_720_fp8_e4m3fn.safetensors">Hunyuan Video FastVideo 720 Fp8 e4m3fn </a>- Place in: ComfyUI_windows_portable\ComfyUI\models\diffusion_modelsDual Clip Loader: <a rel="ugc" href="https://huggingface.co/comfyanonymous/flux_text_encoders/blob/main/clip_l.safetensors">Clip_I.safetensors</a> and <a rel="ugc" href="https://huggingface.co/calcuis/gguf-node/blob/e1363619aea1622debab9520fabf2f4dd7a0977e/llava_llama3_fp8_scaled.safetensors">llava_lhama3_fp8_scaled.safetensors </a>- Place in: ComfyUI_windows_portable\ComfyUI\models\text_encodersVae: <a rel="ugc" href="https://huggingface.co/Kijai/HunyuanVideo_comfy/blob/main/hunyuan_video_vae_bf16.safetensors">hunyuan_video_vae_bf16.safetensors - </a>Place in: ComfyUI_windows_portable\ComfyUI\models\vaeYou can leave everything else as default, or follow the notes. For low VRAM, try decreasing the tile_size and overlap values.

Files with SHA256: f5f1f56e51994c3f2ab529044713a4cdb9952d2898ab3a823c61787960084f3c