Hearthscale/MiniMax-H3-Image-GGUF

MiniMax H3-Image GGUF

GGUF files of the pruned diffusion transformer of MiniMax H3-Image, for stable-diffusion.cpp and Hearthscale. They are converted from minimax_h3_image_pruned_bf16.safetensors in Comfy-Org/MiniMax-H3-Image with the converter of stable-diffusion.cpp master-1088-32d516c, and they load in stable-diffusion.cpp from master-1088.

File	Size	Type of the block weights
minimax_h3_image_pruned-Q8_0.gguf	21.4 GB	Q8_0
minimax_h3_image_pruned-Q5_K_M.gguf	13.9 GB	Q5_K
minimax_h3_image_pruned-Q4_K_M.gguf	11.4 GB	Q4_K
minimax_h3_image_pruned-Q3_K_M.gguf	8.8 GB	Q3_K

In every file, the attention and MLP weights of the 50 blocks and of the token refiner take the type in the name. The AdaLN projections stay in F16, the norms and the condition projection in BF16, and the output heads in F32. The input projection is F32, except in the Q8_0 file, where it is Q8_0. The naming follows leejet/MiniMax-H3-GGUF.

Each file was tested on an NVIDIA RTX 3080 with 10 GB and 32 GB of system memory. These pictures come from the Q4_K_M file at 1024 × 1024, with seed 42, the Q2_K_M text encoder and stable-diffusion.cpp's master-1088-32d516c release build:

A night market food stall in Taipei, a hand-painted menu board reading "BEEF NOODLES 120" and "SCALLION PANCAKE 60", steam rising from a pot, warm lantern light, 35mm photo	Portrait of an elderly fisherman in a yellow raincoat on a harbour wall at dawn, light rain, weathered skin and kind eyes, fishing boats behind him, 85mm photo, shallow depth of field
Concept art of a city built into the ribs of a giant fossil in a desert, caravans with sails crossing the dunes at sunset, lanterns in the windows, matte painting, wide shot	A minimalist poster for a jazz night, a saxophone made of brass leaves on a cream background, the words "HEARTH JAZZ" in bold letters at the top and "FRIDAY 9 PM" at the bottom, Swiss design

Hearthscale

1. Update Hearthscale to 0.3.2, which ships stable-diffusion.cpp master-1088, the first build that reads MiniMax H3-Image.
2. Open the Server, go to Explore, and search for Hearthscale/MiniMax-H3-Image-GGUF.
3. Choose a file. The Server offers the VAE and the text encoder beside it.
4. Draw and edit in Image studio. Each picture takes 6 steps, on the schedule the model was distilled for.

stable-diffusion.cpp

You also need the VAE and the text encoder:

• VAE: vae/minimax_h3_image_vae_fp16.safetensors from Comfy-Org/MiniMax-H3-Image
• Text encoder: qwen3vl_32b_minimax_h3-Q4_K_M.gguf or qwen3vl_32b_minimax_h3-Q2_K_M.gguf from leejet/MiniMax-H3-GGUF

sd-cli --diffusion-model minimax_h3_image_pruned-Q4_K_M.gguf \
  --vae minimax_h3_image_vae_fp16.safetensors \
  --llm qwen3vl_32b_minimax_h3-Q4_K_M.gguf \
  -p "A silver tabby kitten holding a sign reading 'sd.cpp', natural daylight" \
  --cfg-scale 1.0 --sampling-method euler --sigmas minimax-h3-image \
  -W 1024 -H 1024 --offload-to-cpu --diffusion-fa --vae-tiling

To edit a picture, add -r picture.png and say what to change, for example -p "Give her a red knitted beanie. Keep everything else exactly the same.". Repeat -r for more reference pictures. See the MiniMax-H3-Image guide of stable-diffusion.cpp.

Licence

The files follow the licence of the original model, the MiniMax-H3-Image Community License Agreement. Read it before you use them: it limits where the model may be used.