This Lora was trained with 80 videos (portrait ratio) of 24 frames in total each, only double blocks were trained, so it only has 172 MB, It is epoch number 52 out of 100 epochs of total training. I recommend strength 1 to 1.5 If you are using comfyui's native nodes to generate videos using Hunyuan I advise you to test generating with the sampler/scheduler: dpmpp_2m + beta. To use it, you can use the trigger word: 360c4m3r4 or follow a prompt structure that I think works very well, which is the following: 360c4m3r4 The scene features a close-up view of a {CHARACTER DETAILS}, capturing their face and upper body from a low angle. {BACKGROUND DETAILS}. The {CHARACTER} {ACTION}, its {mouth/nose} prominently visible as it moves closer. There are no significant changes in the {CHARACTER} viewing angle during the scene; the focus remains on the {CHARACTER} face throughout. Example:360c4m3r4 The scene features a close-up view of a dog yorkshire terrier, capturing their face and upper body from a low angle. The background is a super market. The dog yorkshire terrier appears to be walking towards the viewer, its mounth prominently visible as it moves closer. There are no significant changes in the dog yorkshire terrier viewing angle during the scene; the focus remains on the dog yorkshire terrier face throughout. Carry out your tests and adjustments and see which prompt best suits your needs. I trained using the fork I made of diffusion-pipe, it has an interface using gradio and a docker image that makes it easy to use, with just one command you have the entire environment set up and the models already downloaded, if you want to use it follow the instructions in the README of the repository: <a target="_blank" rel="ugc" href="https://github.com/alisson-anjos/diffusion-pipe-ui">https://github.com/alisson-anjos/diffusion-pipe-ui</a> or you can use the template for Runpod or VastAI. <a target="_blank" rel="ugc" href="https://runpod.io/console/deploy?template=t46lnd7p4b&amp;ref=8t518hht">https://runpod.io/console/deploy?template=t46lnd7p4b&amp;ref=8t518hht</a> <a target="_blank" rel="ugc" href="https://cloud.vast.ai/?ref_id=142589&amp;creator_id=142589&amp;name=Hunyuan%20Lora%20Train%20Simple%20Interface">https://cloud.vast.ai/?ref_id=142589&amp;creator_id=142589&amp;name=Hunyuan%20Lora%20Train%20Simple%20Interface</a> Article: <a target="_blank" rel="ugc" href="https://civarchive.com/articles/10547/train-lora-for-hunyuan-video-using-diffusion-pipe-interface-with-docker-runpod-and-vastai">Train LoRA for Hunyuan Video using diffusion-pipe Interface with Docker, RunPod, and </a><a target="_blank" rel="ugc" href="http://Vast.AI">Vast.AI</a><a target="_blank" rel="ugc" href="https://civarchive.com/articles/10547/train-lora-for-hunyuan-video-using-diffusion-pipe-interface-with-docker-runpod-and-vastai"> | Civitai</a>

Files with SHA256: d5cac846a48bb4d5b59f432d09cef454833b61782ba185f6d4dc0e56504e5de7