# AI use cases for Kokon (GabrieleLozupone/kokon) Repo: [GabrieleLozupone/kokon](https://github.com/GabrieleLozupone/kokon) | Use cases: 4 | Page: https://dedupedai.com/use-cases/gabrielelozupone-kokon/ Kokon is a free, open source desktop app that edits and creates photos from text descriptions on your own GPU or Mac, powered by Qwen-Image 2.1, with an MCP server for AI agents. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Edit a photo by describing the change in plain words: https://dedupedai.com/use-cases/gabrielelozupone-kokon/describe-edits-to-a-photo/ - Let an AI agent create and edit images on your machine: https://dedupedai.com/use-cases/gabrielelozupone-kokon/agent-driven-image-edits-via-mcp/ - Remove or replace one object without touching the rest: https://dedupedai.com/use-cases/gabrielelozupone-kokon/remove-or-replace-one-object-in-a-photo/ - Upscale and color-grade a result for final delivery: https://dedupedai.com/use-cases/gabrielelozupone-kokon/finish-images-with-upscaling-and-color-grading/ ## How to edit a photo by describing the change in plain words with Kokon Repo: [GabrieleLozupone/kokon](https://github.com/GabrieleLozupone/kokon) | Difficulty: Beginner | Time: 30 minutes | Basis: From the README Install Kokon, let it set up models, then type an edit like 'make it golden hour' and get the result locally. **Who it is for:** A hobbyist or small business owner with a photo to fix, on a Windows PC, Mac or Linux machine with a suitable GPU. **The problem:** Photo editors have sliders and layers to learn, and cloud tools upload personal pictures to someone else's servers. **The outcome:** An edited photo at the original size and aspect ratio, produced entirely on the user's own computer. ### Steps 1. Download the installer for your system from the latest release on the Kokon GitHub page (Windows: the x64 setup .exe; Mac: the Apple Silicon .dmg in beta; Linux: the .deb or .rpm in beta). 2. Run the installer. On Windows it installs for your user only, so administrator rights are not needed. 3. On first launch, let Kokon detect your graphics card or Mac memory, accept the recommended preset and wait for the model download to finish and verify. 4. Drop or paste a photo into Kokon. 5. Describe the change in a sentence, for example 'Make it golden hour', and press Enter. ### Prompt to paste into your agent ```text Use the Kokon app on this machine to edit the photo at with the instruction ''. Save the result next to the original and report the output file path. Check that the output has the same width, height and aspect ratio as the input photo. ``` ### Caveat The first setup downloads about 12 GB and needs a GPU with 12 GB VRAM or an Apple Silicon Mac with 16 GB or more; Mac and Linux builds, AMD and Intel GPU paths are marked beta in the README. 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/gabrielelozupone-kokon/describe-edits-to-a-photo/ ## How to let an AI agent create and edit images on your machine with Kokon Repo: [GabrieleLozupone/kokon](https://github.com/GabrieleLozupone/kokon) | Difficulty: Advanced | Time: 45 minutes | Basis: From the README Connect Claude or Cursor to Kokon's built-in MCP server so the agent can generate and edit images locally. **Who it is for:** A developer or content operator whose AI coding or chat agent needs to produce or revise images as part of a workflow, without sending files to a cloud image service. **The problem:** Agents can write text but cannot make or edit images, and cloud image APIs send prompts and pictures off the machine. **The outcome:** An agent that creates or edits images through Kokon, with the files written to the user's own computer. ### Steps 1. Install Kokon and finish the model setup, following the Install section of the README. 2. Open the AI agents section of the Kokon README to find the MCP server and local HTTP API details (inferred: the README links to this section from its header). 3. Add the Kokon MCP server to your agent's MCP configuration, using the server details from that section. (inferred, check the README) 4. Restart the agent and confirm that the Kokon tools are listed. 5. Ask the agent for one new image and one edit of an existing photo, then open the results in Kokon's gallery to check them. ### Prompt to paste into your agent ```text Using the Kokon MCP server in this repo (GabrieleLozupone/kokon), edit the image at with the instruction '' and save the result to . Expected output: the path of the new image file. Verify that the file exists and that its dimensions match the source image. ``` ### Caveat The README names the MCP server and local HTTP API but does not show their setup commands or tool names in the visible text, so the agent must read the AI agents section before configuring anything. 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/gabrielelozupone-kokon/agent-driven-image-edits-via-mcp/ ## How to remove or replace one object without touching the rest with Kokon Repo: [GabrieleLozupone/kokon](https://github.com/GabrieleLozupone/kokon) | Difficulty: Beginner | Time: 30 minutes | Basis: From the README Paint over an area, pick Remove or Replace, and let the model edit only that region. **Who it is for:** A seller, designer or family photographer with one unwanted object or item to swap in an otherwise good photo. **The problem:** Cleaning up one item by hand in a photo editor takes time, and a full-image AI edit often changes the rest of the picture. **The outcome:** A photo where only the painted area changed, with the rest of the image left as it was. ### Steps 1. Complete the Kokon setup, as described in the Install section. 2. Drop the photo into Kokon. 3. Paint over the area you want to change with the brush. 4. Choose Remove for an object, or Replace and describe the new item, such as 'Replace the yellow armchair with a deep green velvet sofa'. 5. Compare the result with the original using Slider, Flicker or Changes, and undo with Ctrl + Z if needed. ### Prompt to paste into your agent ```text In the Kokon app, remove the object at the painted area of and save the result as . Expected output: a new image file with the same size as the source. Verify by comparing the unpainted parts of the source and result images to confirm they are unchanged. ``` ### Caveat The painted-area workflow depends on the model setup finishing first, and Remove needs no prompt while Replace needs a clear description of the new item. 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/gabrielelozupone-kokon/remove-or-replace-one-object-in-a-photo/ ## How to upscale and color-grade a result for final delivery with Kokon Repo: [GabrieleLozupone/kokon](https://github.com/GabrieleLozupone/kokon) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Take a generated or edited image to 2x, 4K or a custom width, then apply a LUT look. **Who it is for:** A content creator or small shop owner who needs a larger, consistently colored image for a website, print or social post. **The problem:** Plain resizing makes images blurry, and matching a color look across several images is slow in general-purpose editors. **The outcome:** A larger image with an AI upscaler and a chosen color look, saved from Kokon. ### Steps 1. Create or edit an image in Kokon, as described in the Install section. 2. Open the result and choose Upscale, then pick 2x, 4K, the source size or a custom width. 3. Open the color-grade option and pick one of the eight built-in looks, or load your own .cube file. 4. Adjust the LUT intensity with the live preview. 5. Save the result from Kokon. ### Prompt to paste into your agent ```text Using the Kokon app, upscale the image at to pixels wide and apply the color look . Expected output: a saved image file at . Verify that the output width equals the target width. ``` ### Caveat The upscaler model is listed in the README as part of the PC presets, so Mac users should confirm the upscaler is installed before relying on this step. 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/gabrielelozupone-kokon/finish-images-with-upscaling-and-color-grading/