Skip to content
lumalabs.ai
Getting started

Models

Capabilities and pricing for the uni-1 family — uni-1 and uni-1-max (higher-quality output at a higher price).

The uni-1 family handles image generation (text to image) and image editing (modifying existing images with text instructions) through a single API endpoint. Two models share the same wire format and capability set; uni-1-max produces higher-quality output than uni-1 at a higher per-image price. Both models are available to all accounts.

ModelWhen to usePer-image price (text-to-image, 2K)
uni-1Default — fast, photorealistic, multi-panel, text rendering$0.0404
uni-1-maxWhen you want higher-quality output than uni-1 and the higher per-image cost is fine$0.1000

See Pricing for the full per-task table and Provisioned Throughput plans.

Both models support the same capabilities and the same wire format. The only differences are output quality and per-image price.

CapabilityDescription
Text renderingRenders readable text on signs, labels, book covers, and other surfaces. Put exact text in quotes in your prompt.
Spatial reasoningProduces accurate shadows, correct perspective, and physically plausible object layouts.
Cultural stylesUnderstands visual traditions — manga panels with screentone shading, ukiyo-e woodblock prints, film noir lighting, art house cinema framing, and more.
Reference-guided generationAccepts up to 9 reference images via image_ref for text-to-image, or up to 8 for image editing (where the source image occupies its own slot).
Image editingModifies existing images via source — change backgrounds, swap objects, transfer styles while preserving unmentioned parts.
Web search groundingWhen web_search: true, searches the web for visual references before generating.
Multi-panel outputGenerates multi-panel sequences (e.g., storyboards) with consistent style when described in the prompt.
ParameterTypeDefaultDescription
promptstringrequiredText description of the image (1–6,000 characters)
typestring"image""image" for generation, "image_edit" for editing
modelstring"uni-1""uni-1" or "uni-1-max"
aspect_ratiostring | nullnull (model picks)Output dimensions (9 options)
stylestring"auto""auto" or "manga"
output_formatstring | nullnull (model picks)"png" or "jpeg"
web_searchbooleanfalseSearch the web for visual references before generating
image_refarray[]Up to 9 reference images for type: "image", or 8 for type: "image_edit" (URL or base64)
sourceobject | nullSource image for editing (required when type: "image_edit")

For detailed parameter documentation, see Image generation and Image editing.

  • Photorealistic scenes — Natural lighting, material textures, and depth of field
  • Text rendering — Readable text on signs, labels, and surfaces with accurate letterforms
  • Multi-panel consistency — Multiple views of the same scene with consistent perspective
  • Style transfer — Faithfully adopts artistic styles from reference images or text descriptions
  • Spatial precision — Correct object placement, shadows, reflections, and perspective
  • Cultural visual language — Manga, cinematic framing, traditional art styles
ValueOrientationTypical use case
3:1Ultra-wide landscapePanoramic banners
2:1Wide landscapeWebsite headers
16:9Standard widescreenHero images, desktop wallpapers
3:2Classic landscapeTraditional photography
1:1SquareProfile pictures, social media
2:3Classic portraitBook covers, posters
9:16Standard portraitMobile wallpapers, stories
1:2Tall portraitVertical banners
1:3Ultra-tall portraitTall signage
FormatBest for
pngLossless quality
jpegSmaller file size, photographs

When output_format is omitted, the model picks a format based on the prompt.

Generated images are delivered as presigned URLs that expire after 1 hour. Download images promptly or request a fresh URL by polling GET /v1/generations/{id} again.

The default. Use it for the broad set of image-generation and image-editing tasks — see Capabilities above. Defaults to model: "uni-1" when omitted from the request.

See Pricing — uni-1 for the per-task table.

uni-1-max produces higher-quality output than uni-1. Same wire format and parameter set — same prompts, same aspect_ratio, same image_ref semantics. Pass "model": "uni-1-max" on POST /v1/generations.

{ "model": "uni-1-max", "prompt": "A neon-lit Tokyo alley in the rain" }

See Pricing — uni-1-max for the per-task table.

  • Image generation — Full parameter reference
  • Image editing — Modify existing images with text prompts and reference images
  • Pricing — Pay-as-you-go and Provisioned Throughput plans
  • FAQ — Quick answers to common questions
  • API Reference — Complete endpoint specifications