Reve capabilities

Lets users pick the output shape from a set of aspect-ratio presets covering landscape, portrait, photography, and square formats.

An optional toggle that automatically rewrites and enriches the user's typed prompt — adding stylistic detail and visual language — before generation, so short prompts produce richer, more detailed images.

A developer REST API (in beta) for programmatically generating, editing, and remixing images with the Reve model, billed on a per-image credit basis.

Turns generated or uploaded images into short cinematic video clips, with text-to-video, image-to-video (start/end frames), and image-reference modes, plus optional audio.

Renders accurate, legible text and typography directly inside generated images — signs, shirts, newspapers, packaging, labels, menus — with positional control over where text sits in the composition.

The full Reve product runs in a mobile web browser, so users can generate and edit images on a phone by logging into the site — no dedicated native app required.

Combines multiple uploaded reference images into a single new output, and transforms an uploaded image with natural-language instructions, letting users mix subjects, styles, and compositions.

Generates images at native 4K x 4K resolution (true 16 megapixels) in a single pass, suitable for print workflows without a separate upscaling step.

Reve sells a consumer subscription with three tiers — Free, Lite, and Pro — metered by a daily/monthly 'energy' allowance, plus never-expiring Booster Packs and a separately metered video allowance. Users own the rights to images they generate.

An editor surface that lets users edit specific regions of a generated image directly — nudging objects around, recomposing the scene, and making targeted changes as an iterative creative process rather than a one-shot text command.

Generates images from a natural-language text prompt inside the Reve web app, producing detailed, prompt-faithful images across photorealistic and stylized aesthetics.