MiniMax H3
Built on the default MiniMax H3 workflow from ComfyUI. Connect any audio source into the workflow, including audio from another video.
The generated video will fit your description + source audio.
Model Links
nodes
ComfyUI-VideoHelperSuite - https://github.com/kosinkadink/ComfyUI-VideoHelperSuite (for extracting audio from an uploaded video)
[OPTIONAL] ComfyUI-Spectrum-MiniMax-H3 - https://github.com/xmarre/ComfyUI-Spectrum-MiniMax-H3) (MUCH faster generation time, ~3x fast)
vae
minimax_h3_video_vae_fp16.safetensors - https://huggingface.co/Comfy-Org/MiniMax-H3/resolve/main/vae/minimax_h3_video_vae_fp16.safetensors
minimax_h3_audio_vae_fp32.safetensors - https://huggingface.co/Comfy-Org/MiniMax-H3/resolve/main/vae/minimax_h3_audio_vae_fp32.safetensors
diffusion_models
minimax_h3_fl2va_pruned_int8_convrot.safetensors - https://huggingface.co/Comfy-Org/MiniMax-H3/resolve/main/diffusion_models/minimax_h3_fl2va_pruned_int8_convrot.safetensors
text_encoders
qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors - https://huggingface.co/Comfy-Org/MiniMax-H3/resolve/main/text_encoders/qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors
Description
FAQ
Comments (2)
Could this be used to automate music video creation? Have you tried it? I'm guessing in order for it to be practical, it would have to render enough <5 second scenes to match the song length and then have some form of automation to stitch the scenes together. The whole thing could be governed by one of the cool prompt-improver engines that are floating around, which automatically elevate a crappy prompt into the proper, detailed H3 formatting.
You still need to include the words of the audio in the prompt, so I would rig a transcription to description pipeline and yeah I think it's doable.