# AI use cases for Comfy UI (Comfy-Org/ComfyUI) Repo: [Comfy-Org/ComfyUI](https://github.com/Comfy-Org/ComfyUI) | Use cases: 4 | Page: https://dedupedai.com/use-cases/comfy-org-comfyui/ ComfyUI is an open-source node graph GUI, API, and backend for running diffusion models locally or in the cloud to generate images, video, audio, and 3D. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Generate images from ready-made template workflows: https://dedupedai.com/use-cases/comfy-org-comfyui/generate-images-with-template-workflows/ - Run image and video generation fully offline: https://dedupedai.com/use-cases/comfy-org-comfyui/run-offline-generation-for-private-work/ - Turn a complex workflow into a simple app or API: https://dedupedai.com/use-cases/comfy-org-comfyui/expose-a-workflow-as-an-app-or-api/ - Run large open-source models on a modest GPU: https://dedupedai.com/use-cases/comfy-org-comfyui/run-large-models-on-modest-hardware/ ## How to generate images from ready-made template workflows with Comfy UI Repo: [Comfy-Org/ComfyUI](https://github.com/Comfy-Org/ComfyUI) | Difficulty: Beginner | Time: 30 minutes | Basis: From the README Install the desktop app, open a template, and queue your first image from a visual node graph. **Who it is for:** A visual artist or marketer with no coding background who needs to produce images with open-source models on a personal computer. **The problem:** Open-source image models are hard to set up and hard to control without a clear, visual way to change prompts, models, and settings. **The outcome:** A generated image from a template workflow, plus a saved workflow file that can be reused and changed. ### Steps 1. Download the Desktop Application from https://www.comfy.org/download (available on Windows and macOS). 2. Open a template from https://comfy.org/workflows and load it in Comfy UI. 3. Change the prompt text in the text input node and set the model loader to a model you have installed. 4. Press Ctrl + Enter to queue the current graph for generation. 5. Press Ctrl + S to save the workflow as JSON so you can reopen it later. ### Prompt to paste into your agent ```text Using the ComfyUI repo at https://github.com/Comfy-Org/ComfyUI, load the template workflow for from comfy.org/workflows, set its prompt to , queue it, and save the result. Expected output: one generated image file and a saved workflow JSON. Verify that the output image opens and matches the prompt. ``` ### Caveat The README lists the model families supported but does not say which model files come bundled, so you may need to download a checkpoint yourself before the first run. Risk, reverse engineering: Generated images of real people or copyrighted characters can cause legal or consent problems, so check the rights and model license before publishing output. Dual-use notice: only use this on software, sites and systems you own or are authorized to test. 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/comfy-org-comfyui/generate-images-with-template-workflows/ ## How to run image and video generation fully offline with Comfy UI Repo: [Comfy-Org/ComfyUI](https://github.com/Comfy-Org/ComfyUI) | Difficulty: Intermediate | Time: 90 minutes | Basis: From the README Install Comfy UI locally, start it with --offline, and keep generation off external services. **Who it is for:** A studio or freelancer handling confidential client assets who must keep prompts, images, and models on their own machine. **The problem:** Cloud generation services send prompts and outputs to third parties, which may break client confidentiality rules. **The outcome:** A local Comfy UI session that generates media without calling the paid API nodes or downloading anything you did not request. ### Steps 1. Follow the Manual Install section of the README (https://github.com/Comfy-Org/ComfyUI#manual-install-windows-linux) for your operating system and GPU type. 2. Start Comfy UI from the install folder with python main.py --offline. (inferred, check the README) 3. Place your model files in the model folders, or point to them by copying extra_model_paths.yaml.example to extra_model_paths.yaml and editing the paths. 4. Build or load a workflow from the local template library and queue it with Ctrl + Enter. 5. Confirm the paid API nodes are not used, then export results from your local output folder. ### Prompt to paste into your agent ```text Using the ComfyUI repo at https://github.com/Comfy-Org/ComfyUI, start ComfyUI in offline mode on and run the workflow with the model stored at . Expected output: generated media files saved locally with no network calls to paid API nodes. Verify that the server log shows offline mode and that no outbound requests were made during the run. ``` ### Caveat The README says core does not download anything unless requested, but you must still check each custom node you install, since custom nodes are separate code and the README does not say what they do with the network. Dual-use notice: only use this on software, sites and systems you own or are authorized to test. 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/comfy-org-comfyui/run-offline-generation-for-private-work/ ## How to turn a complex workflow into a simple app or API with Comfy UI Repo: [Comfy-Org/ComfyUI](https://github.com/Comfy-Org/ComfyUI) | Difficulty: Advanced | Time: 120 minutes | Basis: From the README Use App Mode and the local API to run a saved workflow from a form or a production pipeline. **Who it is for:** A developer or operations lead who wants non-technical teammates or a pipeline to run a fixed generation workflow without opening the node graph. **The problem:** Complex node graphs are hard for people who only need to change one or two inputs, and pipelines need a stable way to call the same workflow. **The outcome:** A workflow that is available through App Mode for simple use and through the local API for automated runs. ### Steps 1. Open the workflow you want to share and save it with Ctrl + S as JSON. 2. Enable App Mode and choose which inputs (such as prompt and seed) appear in the simple UI. 3. Start Comfy UI locally with python main.py and confirm the local API is reachable on the server address shown in the log. (inferred, check the README) 4. Send the saved workflow to the local API from your application, changing only the inputs you exposed. 5. Collect the output files and check them against a test prompt before you hand the app to users. ### Prompt to paste into your agent ```text Using the ComfyUI repo at https://github.com/Comfy-Org/ComfyUI, save the workflow as JSON, expose its prompt input through App Mode, and submit it to the local API with the prompt . Expected output: one generated file returned by the API and the saved workflow JSON. Verify the API response status and that the output matches the prompt. ``` ### Caveat The README mentions a local API and API endpoints for production pipelines but does not document the endpoint paths or request format, so check the official docs before you write the integration. Dual-use notice: only use this on software, sites and systems you own or are authorized to test. 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/comfy-org-comfyui/expose-a-workflow-as-an-app-or-api/ ## How to run large open-source models on a modest GPU with Comfy UI Repo: [Comfy-Org/ComfyUI](https://github.com/Comfy-Org/ComfyUI) | Difficulty: Intermediate | Time: 120 minutes | Basis: From the README Use Comfy UI's memory management and quantized model support to run bigger models on a small GPU. **Who it is for:** An independent creator with a consumer GPU who wants to try large video or image models without buying new hardware. **The problem:** Large diffusion models usually need more VRAM than a consumer card provides, so they fail to load or run very slowly. **The outcome:** A large model running locally through Comfy UI with weight streaming and offloading, generating output on a small GPU. ### Steps 1. Install Comfy UI using the Manual Install section or the desktop app, depending on your system. 2. Download a quantized version of the model you want, since the README lists support for quantized models. 3. Load the model with a workflow template from comfy.org/workflows that matches your model family. 4. Queue the workflow with Ctrl + Enter and watch the GPU usage while it runs. 5. If the run fails, lower the resolution or frame count and queue it again. ### Prompt to paste into your agent ```text Using the ComfyUI repo at https://github.com/Comfy-Org/ComfyUI, load a quantized version of on a machine with VRAM and generate . Expected output: one generated file and a log showing the peak VRAM used. Verify the run finishes without an out-of-memory error. ``` ### Caveat The README's claim of running the biggest models on 4GB VRAM plus 8GB RAM is the project's own statement, and actual speed and fit depend on the model and settings. Dual-use notice: only use this on software, sites and systems you own or are authorized to test. 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/comfy-org-comfyui/run-large-models-on-modest-hardware/