Make a 16 megapixel picture from a prompt, or edit a picture in plain words. For an edit, you can give MiniMax H3 Image up to seven more pictures and tell it what to take from each one.

How to use it

1. Drop the workflow into ComfyUI and press Run. It opens with a full prompt, so you get a picture right away.

2. For your own picture, write a full description. Or turn Auto prompt on and write a few words: a writer turns them into the long description the model wants.

3. To edit, turn Edit a picture on, load your picture and write the change, for example “Make it snow.”

4. For a reference, connect an image loader to picture 2 on Write the prompt. Add more in order through picture 8. Name the picture in your request, for example “Put the jacket from picture 2 on the woman in picture 1.”

5. Size picks 4 MP or 16 MP. Fast runs 2 steps instead of 6. Upscale the result doubles a 4 MP picture with SeedVR2.

What you need

Update ComfyUI and AusBoss nodes first. MiniMax H3 Image support was added to ComfyUI on 2026-11-10, after version 0.44.0, so you need a build newer than that. This version ran on AusBoss nodes 2.22.0. The workflow opens on High VRAM.

• High VRAM · minimax_h3_image_pruned_int8_convrot.safetensors → models/diffusion_models · 21.0 GB

• Low VRAM · minimax_h3_image_pruned_w4a8.safetensors → models/diffusion_models · 12.5 GB

• Both · qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors → models/text_encoders · 15.7 GB

• Both · minimax_h3_image_vae_fp16.safetensors → models/vae · 1.2 GB

• Auto prompt · qwen3.5_9b_w4a8_gptq.safetensors → models/text_encoders · 7.3 GB

• Fast · minimax_h3_pdmd_2nfe_comfyui.safetensors → models/loras · 1.96 GB

• 2× upscale · seedvr2_3b_int8_convrot.safetensors → models/diffusion_models · 3.46 GB

• 2× upscale · seedvr2_ema_vae_fp16.safetensors → models/vae · 501 MB

Tips

• On an RTX 5090 a new 16 MP picture took about 22 seconds and a 16 MP edit about 35. At 4 MP an edit took about 12.

• Put any words that should appear in the picture in quotes. Signs and posters spell them right. Small text far from the subject can still come out as nonsense.

• Leave Auto prompt off when your prompt is already a full description. In our runs of the opening prompt, rewriting it tangled the hands.

• For an edit, say what should stay the same as well as what you want to change.

• Fast frames a little tighter and small words can come out wrong. Leave it off for posters and covers.

MiniMax-H3-Image is by MiniMax and uses the MiniMax-H3-Image Community License. Powered by MiniMax-H3-Image. The 2-step LoRA is akhaliq's pDMD. Optional upscaling uses SeedVR2 under Apache 2.0.

v1.0 (2026-11-10) · First release.

Report an issue · Discord · YouTube @ausbossdev · X @Zanzibased · GitHub