Generate ComfyUI Workflow

Here's a workflow that allows you to ask a question, generate an answer, and then create an image based on that answer using the SDXL model: { "1": { "inputs": { "text": "What is a unique animal hybrid that doesn't exist in real life?", "max_tokens": 50, "preset": "None" }, "class_type": "CLIPTextEncode", "_meta": { "title": "Question Input" } }, "2": { "inputs": { "text": [ "1", 0 ] }, "class_type": "moondream_text_model", "_meta": { "title": "Moondream Text Model" } }, "3": { "inputs": { "ckpt_name": "albedobaseXL_v12.safetensors" }, "class_type": "CheckpointLoaderSimple", "_meta": { "title": "Load SDXL Checkpoint" } }, "4": { "inputs": { "width": 1024, "height": 1024, "batch_size": 1 }, "class_type": "EmptyLatentImage", "_meta": { "title": "Empty Latent Image" } }, "5": { "inputs": { "text": [ "2", 0 ], "clip": [ "3", 1 ] }, "class_type": "CLIPTextEncode", "_meta": { "title": "CLIP Text Encode (Prompt)" } }, "6": { "inputs": { "text": "low quality, blurry, distorted", "clip": [ "3", 1 ] }, "class_type": "CLIPTextEncode", "_meta": { "title": "CLIP Text Encode (Negative Prompt)" } }, "7": { "inputs": { "seed": 721897303308196, "steps": 50, "cfg": 8, "sampler_name": "dpmpp_sde", "scheduler": "karras", "denoise": 1, "model": [ "3", 0 ], "positive": [ "5", 0 ], "negative": [ "6", 0 ], "latent_image": [ "4", 0 ] }, "class_type": "KSampler", "_meta": { "title": "KSampler" } }, "8": { "inputs": { "samples": [ "7", 0 ], "vae": [ "3", 2 ] }, "class_type": "VAEDecode", "_meta": { "title": "VAE Decode" } }, "9": { "inputs": { "filename_prefix": "question_answer_image", "images": [ "8", 0 ] }, "class_type": "SaveImage", "_meta": { "title": "Save Image" } } } This workflow does the following: 1. It starts with a "Question Input" node where you can input your question. 2. The question is then passed to a "Moondream Text Model" which generates an answer based on the question. 3. The generated answer is used as the prompt for image generation. 4. The workflow uses the SDXL model (specifically albedobaseXL_v12) to generate the image. 5. The generated image is then saved. To use this workflow: 1. Input your question in the "Question Input" node. 2. The Moondream model will generate an answer. 3. This answer will be used as the prompt for the SDXL model to generate an image. 4. The resulting image will be based on the answer to your question. This workflow allows for a creative process where you can ask a question and get a visual representation of the answer.