# AI use cases for Comfy UI Mini Max H3 Image Studio (astropuzzo/ComfyUI-MiniMax-H3-Image-Studio) Repo: [astropuzzo/ComfyUI-MiniMax-H3-Image-Studio](https://github.com/astropuzzo/ComfyUI-MiniMax-H3-Image-Studio) | Use cases: 3 | Page: https://dedupedai.com/use-cases/astropuzzo-comfyui-minimax-h3-image-studio/ A ComfyUI custom node pack with two starter workflows for MiniMax H3 text-to-image generation and image editing, with optional stackable LoRAs. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Generate a still image from a text prompt: https://dedupedai.com/use-cases/astropuzzo-comfyui-minimax-h3-image-studio/generate-still-image-from-text-prompt/ - Edit an existing image with a text instruction: https://dedupedai.com/use-cases/astropuzzo-comfyui-minimax-h3-image-studio/edit-existing-image-with-text-instruction/ - Combine multiple reference images in one edit: https://dedupedai.com/use-cases/astropuzzo-comfyui-minimax-h3-image-studio/combine-multiple-reference-images-in-one-edit/ ## How to generate a still image from a text prompt with Comfy UI Mini Max H3 Image Studio Repo: [astropuzzo/ComfyUI-MiniMax-H3-Image-Studio](https://github.com/astropuzzo/ComfyUI-MiniMax-H3-Image-Studio) | Difficulty: Intermediate | Time: 90 minutes | Basis: From the README Render a single still image from a written prompt using the MiniMax H3 Generate workflow in ComfyUI. **Who it is for:** A ComfyUI user who needs a quick still image for a draft, mockup or concept and already has the H3 models on disk. **The problem:** Setting up H3 image generation by hand means wiring models, prompt, sampler and save nodes correctly. **The outcome:** A saved still image rendered from a prompt through the Generate template. ### Steps 1. Install Image Studio through ComfyUI Manager or clone the repository into ComfyUI/custom_nodes/, then restart ComfyUI and refresh the browser. 2. Place the three starting model files in models/diffusion_models/, models/text_encoders/ and models/vae/ as listed in the README. 3. Open Templates, choose MiniMax H3 Image Studio, and load the Generate workflow. 4. In H3 • Models, select the three files, write the prompt, choose aspect ratio and megapixels (3 MP is the starting point), then queue. 5. Check the Save Image output and adjust seed or steps in Render Image if needed. ### Prompt to paste into your agent ```text Using the astropuzzo/ComfyUI-MiniMax-H3-Image-Studio repo, set up the Generate workflow in ComfyUI for the prompt . Confirm ComfyUI is 0.37.0 or newer and the three model files are in their folders, then queue the job. Expected output: one saved image from Save Image; verify that the run_info in Render Image lists the applied settings. ``` ### Caveat The README says the model is experimental, so prompt adherence, identity, typography and fine detail vary between runs. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/astropuzzo-comfyui-minimax-h3-image-studio/generate-still-image-from-text-prompt/ ## How to edit an existing image with a text instruction with Comfy UI Mini Max H3 Image Studio Repo: [astropuzzo/ComfyUI-MiniMax-H3-Image-Studio](https://github.com/astropuzzo/ComfyUI-MiniMax-H3-Image-Studio) | Difficulty: Intermediate | Time: 90 minutes | Basis: From the README Change a source image, such as its background, by giving H3 a plain-language instruction in the Edit workflow. **Who it is for:** A creator with a photo or render that needs a targeted change, like swapping the background, and who wants to avoid redrawing it manually. **The problem:** Retouching a source image by hand takes time, and a generic generator may change parts of the image that should stay the same. **The outcome:** An edited image in which the requested change is applied while the source is used as the base. ### Steps 1. Install Image Studio and place the three starting model files as described in the README. 2. Open the Edit template, which adds a Load Image node connected to Image Edit. 3. Load the source image into Load Image and write an ordinary instruction, such as one that replaces the background and keeps the subject unchanged. 4. Keep Edit aspect ratio set to source image, select the output size, then queue the workflow. 5. Review the result and rerun with a changed seed or steps if the edit missed the instruction. ### Prompt to paste into your agent ```text Using the astropuzzo/ComfyUI-MiniMax-H3-Image-Studio repo, run the Edit workflow in ComfyUI on the source image with the instruction . Expected output: one edited image from Save Image that keeps the source proportions. Verify that the subject and layout the instruction says to keep are still present. ``` ### Caveat The README warns that the model can still miss an instruction, so successful edits are not guaranteed. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/astropuzzo-comfyui-minimax-h3-image-studio/edit-existing-image-with-text-instruction/ ## How to combine multiple reference images in one edit with Comfy UI Mini Max H3 Image Studio Repo: [astropuzzo/ComfyUI-MiniMax-H3-Image-Studio](https://github.com/astropuzzo/ComfyUI-MiniMax-H3-Image-Studio) | Difficulty: Advanced | Time: 120 minutes | Basis: From the README Assign roles to up to nine pictures so H3 takes a person, pose and room from separate references. **Who it is for:** A designer or content maker who has several reference pictures and needs one composite image that uses a specific element from each. **The problem:** Merging elements from several photos by hand in an editor is slow, and a single-source prompt cannot name which picture supplies which element. **The outcome:** One composite image built from the roles assigned to each reference picture. ### Steps 1. Set up Image Studio and the three model files as described in the README. 2. Open the Edit workflow and add picture sockets to Image Edit, which grow as needed up to nine pictures total. 3. Load each reference image into its own socket, noting which picture is which. 4. Write a prompt that assigns each picture an explicit role, for example keeping the person and clothes from one picture and using the pose from another. 5. Queue the workflow and check that each role was applied, then rerun with a different seed if needed. ### Prompt to paste into your agent ```text Using the astropuzzo/ComfyUI-MiniMax-H3-Image-Studio repo, run the Edit workflow in ComfyUI with the reference images and the role prompt . Expected output: one composite image from Save Image. Verify that each reference contributed the role the prompt assigned to it. ``` ### Caveat Multi-reference results vary and the README does not guarantee that the model follows every role assignment. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/astropuzzo-comfyui-minimax-h3-image-studio/combine-multiple-reference-images-in-one-edit/