Turn a short prompt into a detailed image
Open this ready-to-run workflow on Floyo. No install needed.
HOW IT WORKS
Step 1. Write your prompt. A one-line idea is enough, like "an abandoned Victorian mansion overtaken by vines, oil painting style." The enhancer fills in the lighting, mood, and composition. Works great with: posters · infographics · concept art · illustrations
Step 2. Choose enhancer on or off. Leave it on to let short prompts do more work. Turn it off when you want your exact wording to drive the image.
Step 3. Hit run and download. You get back a 1024px image. Preview it in the workflow, then download. Ready for: Photoshop · Figma · Canva · any editor
First time? Leave every setting as-is. The defaults (1024 x 1024 · 20 steps · CFG 4 · enhancer on) are the right starting point for almost everyone.
Overview
This workflow generates an image from a text prompt using ERNIE Image, an open-weight model from Baidu. It runs two models, not one. A small prompt enhancer rewrites your input into a longer, structured visual description, then the 8B image model renders it. So a one-line idea comes back as a detailed image without you writing a paragraph. The enhancer also reads your width and height, so it shapes its description to match your aspect ratio. ERNIE Image is unusually good at legible in-image text, which makes it a fit for posters, signage, and layout work where most image models produce garbled lettering. You type a prompt, hit run, and get the image back in about 37 seconds. No setup, no nodes to wire.
Who it's for: creators who want detailed images without writing long prompts, and anyone who needs readable text inside the image. Not for: image editing or character consistency across shots. Use a Qwen Image Edit or FLUX.2 Klein workflow for those.
Why Floyo
Floyo is the only ComfyUI platform built for teams in the browser.
Made for teams. Share run history, files, and models across your whole team. A teammate opens your exact run and picks up where you left off. No file handoffs, no version confusion.
No install, no setup. Every workflow and model is preloaded. Open it in your browser and run. Nothing to download, nothing to configure.
No local hardware. Workflows run on H100 NVL GPUs, so heavy models run fast without a card of your own. Your VRAM stops being the limit.
Open and closed models in one place. Floyo runs open-source workflows and API models side by side.
How to use (in your browser on Floyo)
Type your prompt. Keep it short and let the enhancer expand it, or turn the enhancer off and write it in full.
Add a negative prompt if you want, then run with the defaults (1024 x 1024, 20 steps, CFG 4).
Download the image from the run, ready for any editor.
Expectations The models are preloaded, so there is nothing to download. Running it needs a free Floyo account. Each image comes back in about 37 seconds. One tradeoff to know: with the enhancer on you get more detail but less control over phrasing, so switch it off when the exact wording matters. Fix the seed to compare prompt variations on the same composition, and keep CFG near 4, since pushing it to 6 and above can cause burn and color artifacts. ERNIE Image is Apache 2.0 licensed, so both the model and your images are yours to use commercially.
Use Cases
Posters & Text-Heavy Design. Generate images with readable in-image text for posters, signage, infographics, and ad frames.
Concept Art & Mood Boards. Type a rough idea and get a fully described image, useful for early exploration before you know what you want.
Illustrations & Scenes. Produce general scenes and illustrations across photoreal, design, and stylized looks.
Fast Ideation. Fix a seed, change one line, and compare prompt variations on the same composition.
What can ERNIE Image generate?
