Prompt Library

It is a free, searchable collection of thousands of ready-made AI image prompts and AI video prompts, each shown next to the picture or clip it produced. Instead of starting at a blank box, you can browse real examples, find a look you want, and copy the exact text that made it.

Yes. The whole prompt library is free to browse, search and copy — no account, no signup and no credit card. You only need a Fulhar account if you want to generate images or video on Fulhar itself, which runs on credits.

A prompt is plain text, so you can paste it into any text-to-image or text-to-video generator — Midjourney, ChatGPT and DALL·E, Google Gemini and Veo, Sora, Stable Diffusion, Flux, Leonardo, Kling, Runway or the generators built into Fulhar. Models read wording slightly differently, so a prompt written for one may need small edits on another. Where we know which model made an example, it is shown as a badge on the prompt's page.

Open any card, press Copy prompt, then paste it into your AI generator and run it. From there, change one thing at a time — the subject, the lighting, the camera angle, the art style — and re-run. Editing a prompt that already works is far quicker than writing one from nothing.

The library is organised into portraits, cinematic shots, ads and product photography, brand and logo design, posters and visuals, illustration, wallpapers, photography, design, 3D, storyboards and video prompts. Use the category tabs to narrow the grid, or the search box to look inside the prompt text itself.

Type into the search box and the library matches against the full prompt text, not just titles — so searching for a style, a lighting term, a camera or a colour palette ("golden hour", "85mm", "neon cyberpunk", "minimal white background") finds every prompt that mentions it. Combine it with a category tab to narrow the results further.

Most strong prompts name six things: the subject, the setting, the art style or medium, the lighting, the camera or composition, and the mood. Vague prompts give generic results, and detail is what separates them — which is the fastest way to learn from this library, since every prompt here is shown beside the image it actually produced.

Yes — every prompt page has a download button for the picture or clip shown with it. Treat those examples as reference and inspiration: they were made by their original creators, so use them to judge what a prompt does rather than publishing them as your own work.

The prompt text is a description of an image, and you are free to reuse and rewrite it. What you may do with the result is set by whichever generator you run it through — Midjourney, OpenAI, Google and the rest each have their own commercial-use terms, and those are what apply to your output. Check the terms of the tool you generate with before using anything commercially.

New prompts are imported in batches as they are collected, and the counts on the category tabs update with them, so the number beside each tab is always the live total rather than a figure typed into the page.