Use case

AI visuals for documentary and faceless YouTube videos

Documentary, explainer and faceless channels need a steady stream of images that match the narration. Melia turns your shot list into a batch so the visual work no longer takes longer than the edit.

The workflow

Plan shots. From your narration, write one image prompt per moment you need to cover. This is your decision — Melia does not read or write the script for you.
Batch generate. Put the prompts in one file, choose an output folder and run. Melia works through the list with parallel workers and retries failures.
Review. Check each result in the Gallery. If something is wrong, adjust that prompt and run again.
Edit. Numbered files (0001.png, 0002.png …) keep the story order when you import them into your editor.

Keeping a consistent look

Melia 1.0 does not include a style library. The reliable approach is to put the same style phrase in every prompt — for example the era, lighting, lens and colour mood — and keep each prompt self-contained. Test on a few prompts first and refine the phrase before the full run.

A real example

These images came out of Melia 1.0 batches with Gemini (Nano Banana 2), unedited:

Himalayan monastery on a cliff above a fog-filled valley at dawn — Documentary — monastery above the fog

The Roman Forum at dusk with marble columns and long golden shadows — History — the Roman Forum at dusk

Melia Gallery showing the three images of the run — The same run in Melia's Gallery, where you review each result before editing.

Costs and limits

Melia is a one-time purchase with no AI credits. Image generation is billed by Google or OpenAI to your own account, so the cost depends on the model and the number of images. Melia produces still images only — it does not generate video, voice-over or music.

Read the bulk-prompt guide for the step-by-step process.