Skip to content
View original post on X: ModelScopeOfficial
Published
26/100

DiffSynth-Studio offers quantized MiniMax-H3-Image checkpoints for low-VRAM inference

AIDiffSynth-Studio releases EntroPack checkpoints for MiniMax-H3-Image, with a bf16 pruned model for 80GB VRAM, an INT8 model for 24GB, and a 3-bit model for 8GB. ComfyUI workflows are also available in the model repository.
ModelScopeVerified on XOfficial
@ModelScope2022
View on X

Deploy MiniMax-H3-Image on ANY hardware with ultra-low cost! 🤖 https://modelscope.ai/models/DiffSynth-Studio/MiniMax-H3-Image-EntroPack

Fast, low-cost inference is now possible with EntroPack. Choose your checkpoint:
🔹 80GB VRAM: bf16_table_pruned.safetensors (Fully lossless pruned model)
🔹 24GB VRAM: dit_8bit_int8.safetensors (Near-lossless INT8 + ConvRot quantization)
🔹 8GB VRAM: dit_extreme_3bpp.safetensors (Extreme 3-bit quantized model)

ComfyUI workflows also available in the model repository. Try it now!

Source details

ModelScope · x.comPublished · added here

Original title: Deploy MiniMax-H3-Image on ANY hardware with ultra-low cost! 🤖 https://modelscope.ai/models/DiffSynth-Studio/MiniMax-H3-Image-EntroPack

Topics

Image generationDeployment & engineeringChina AIProduct updates