ComfyUI Usage & Node Requirements
To run these GGUF models in ComfyUI, install the following custom node packages:
GGUF Loader Node :
ComfyUI-GGUF (GitHub) or install via ComfyUI Node Registry: comfyui-gguf-reboot
Used to load the quantized
.ggufUNet/diffusion models.
MiniMax H3 Turbo Sampler:
ComfyUI-MiniMax-H3-Turbo (GitHub) or install via ComfyUI Node Registry: comfyui-minimax-h3-turbo
Provides the dedicated Turbo sampling schedules and shift calculations for 4-step generation.
Model Placement
Place the downloaded GGUF files in your ComfyUI models directory:
ComfyUI/models/diffusion_models/minimax_h3_fl2v_turbo_4step_v1.0_768p_Q8_CR.ggufShift & Sampling Schedule
Recommended inference step count: 4 steps.
Minimax H3 Video shift: 6, Audio shift: 3
Minimax H3 Turbo Sampler
simple scheduler
Supported Input Modes (FL2VA)
Text-to-Audio-Video (T2VA): No image input connected.
Image-to-Audio-Video (I2VA): One image connected to
first_frame.First-and-Last-Frame-to-Audio-Video (FL2VA): Two images connected to
first_frameandlast_frame.
ComfyUI Workflows
Example workflows are provided
workflows/minimax_h3_t2v-turbo-gguf.json— Text-to-Audio-Video (T2VA) Turbo workflow.workflows/minimax_h3_i2v-turbo-gguf.json— Image-to-Audio-Video (I2VA / FL2VA) Turbo workflow.
License
MiniMax H3 is released under the MiniMax H3 Community License Agreement.
Description
FAQ
Comments (8)
Will you make ref version?
The turbo LoRA is made my lightx2v, they marked the Fl2V version as v1.0 but the ref version is v0.1 so I wanted to wait until they fully train the Ref2V model, but sure why not. I will need day or two though to get everything downloaded again.
Uploaded (Q8_0 finishing uploading in ~10m) but I do not think the quality great.
@molbal the R2V lightx2v turbo lora 4 step v0.1 by kijai work perfectly.
@Pat3dx That's exactly the one I converted :) Perhaps I had a few unlucky generations
bro please make a ref2vid version with larry lora, this work perfect larryvrh/MiniMax-H3-Turbo-Lora · Hugging Face
@one_AI_lover_More sure, I'll get to it might take a few days
@one_AI_lover_More Hey, I took a look now, thanks for the patience. I cannot directly merge it, because it uses a custom node (here: https://github.com/Larryvrh/ComfyUI-MiniMax-H3-Turbo) to apply the lora, BUT you can hook that up to to both the gguf loader node and it should work.