Sign In

Minimax H3 Image Studio 8k+ (12+ GB VRAM)

Updated: Nov 19, 2026

tool

Download

1 variant available

Config Other

Minimax_H3_Image_Studio_8k.json

118.19 KB

Verified:

Type
Workflows
Stats

123

Reviews
Published

Nov 18, 2026

Base Model

MiniMax H3

Hash
AutoV2E8138A18D2

The MiniMax H3 Community License Agreement’s grant excludes the European Union, the United Kingdom, the Republic of Korea and the United States of America.

MiniMax H3

MiniMax_H3_Image_00016_ (1).png

V2.0 (Adaptation for official H3 Image)

MiniMax H3 Image Studio – 8K+ (Ref2VA + Turbo Mode + SeedVR2)

Text-to-image / reference-based image generation with MiniMax H3 image. Includes an optional SeedVR2 4K+ refine pass for details at lower resolutions, in cases where there is insufficient video memory for native generation at high resolution.

Features

- 6-step Turbo sampling (official distill)

- Up to 16 reference images, each in its own group, easy to mute/unmute

- Reference ON/OFF switch, works as plain text-to-image when disabled

- Resolution presets with aspect ratio selection (up to 50+ MP)

- Automatic best-frame selection from the decoded latent

- Optional SeedVR2 (7B GGUF) detail upscale to 8192 px with before/after comparer

- Separate Save nodes for original and upscaled images

How to use

1. Write your prompt in the "Text (Prompt)" node: scene, subject, lighting, style. No camera or sound directions needed, this is a still image.

2. With references: unmute the "Ref Img" nodes you need, load your images and set the switch to true. Without references: set the "Image Reference (On & OFF)" switch to false.

3. Pick aspect ratio and resolution in H3ImageResolutionPreset.

4. Run. If you don't need the upscale, mute the SeedVR2 group.

Required models

Diffusion model: minimax_h3_image_pruned_int8_convrot.safetensors (21 GB)

Text encoder: qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors (14.9 GB)

VAE: minimax_h3_image_vae_fp16.safetensors (1.2 GB)

Upscale (optional): seedvr2_ema_7b-Q4_K_M.gguf + ema_vae_fp16.safetensors

Links:

https://huggingface.co/Comfy-Org/MiniMax-H3-Image/resolve/main/diffusion_models/minimax_h3_image_pruned_int8_convrot.safetensors

https://huggingface.co/Comfy-Org/MiniMax-H3/resolve/main/text_encoders/qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors

https://huggingface.co/Comfy-Org/MiniMax-H3-Image/resolve/main/vae/minimax_h3_image_vae_fp16.safetensors

Custom nodes

- ComfyUI-MiniMax-H3-Image-Studio (astropuzzo)

- rgthree-comfy

- comfyui-zip-ties

- ComfyUI-SeedVR2_VideoUpscaler (for the upscale group)

Tips

- Default is 4-16 MP (for this model), but the architecture itself allows for setting higher values, which is heavy. If you run out of VRAM, lower the preset to 4 MP or less. SeedVR2 can still bring it up to 8-24K (with tile mode).

- SeedVR2 uses block swapping to save VRAM: slower, but works on smaller cards.

- LoRAs from a video model are not supported

- Update ComfyUI to the latest version before use.

Credits

MiniMax H3 / Comfy-Org, astropuzzo (Image Studio nodes), rgthree, SeedVR2 authors.

Tested on: [RTX 4070 / Lastest ComfyUI version]. Generation time: [250-300 sec @ 35MP without SeedVR2 Refine].