Skip to main content
ModelRunner
luma avatar

Luma Uni-1 Max API

luma/uni-1-max

Generate a high-fidelity image from a text prompt, with optional manga styling, web-grounded references, and up to nine reference images to steer composition.

0.102
OpenAPI

Model Input

Input

The text prompt describing the image to generate.

The aspect ratio of the generated image.

Additional Settings

Customize your input with more control.

The visual style of the generated image.

The output image format.

If true, the model may consult the web while generating.

You need to be logged in to run this model and view results.
Log in

Model Output

Output

Generated image output
Generated in 63.74 seconds
Logs (1 lines)

Model Example Requests

Examples

Example output 1Example output 2Example output 3Example output 4

Luma Uni-1 Max API

Luma Uni-1 Max is a text-to-image AI model by luma. On ModelRunner it runs through a REST API or via MCP from any AI assistant, at $0.102 per image.

POST https://queue.modelrunner.run/luma/uni-1-max

cURL

# Submit a request to the queue. Input fields go at the top level of the
# body. The optional reserved "metadata" object holds your own flat string
# tags — stored on the request, never sent to the model; filter later with
# GET https://queue.modelrunner.run/requests?metadata=<url-encoded JSON>.
curl -X POST https://queue.modelrunner.run/luma/uni-1-max \
  -H "Authorization: Key $MRUN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "style": "auto",
    "prompt": "a lone lighthouse on a rocky cliff at golden hour, dramatic clouds, cinematic wide shot",
    "aspect_ratio": "16:9",
    "output_format": "jpeg",
    "enable_web_search": false,
    "reference_image_urls": null,
    "metadata": {
      "project": "my-project"
    }
  }'
# → { "request_id": "...", "status_url": "...", "response_url": "..." }

# Poll status_url until "COMPLETED", then fetch the result
curl "https://queue.modelrunner.run/luma/uni-1-max/requests/$REQUEST_ID/status" \
  -H "Authorization: Key $MRUN_API_KEY"
curl "https://queue.modelrunner.run/luma/uni-1-max/requests/$REQUEST_ID" \
  -H "Authorization: Key $MRUN_API_KEY"

JavaScript

import { modelrunner } from "@modelrunner/client";

const result = await modelrunner.subscribe("luma/uni-1-max", {
  input: {
    "style": "auto",
    "prompt": "a lone lighthouse on a rocky cliff at golden hour, dramatic clouds, cinematic wide shot",
    "aspect_ratio": "16:9",
    "output_format": "jpeg",
    "enable_web_search": false,
    "reference_image_urls": null
  },
});
console.log(result);

Python

import os
import requests

headers = {"Authorization": f"Key {os.environ['MRUN_API_KEY']}"}

submitted = requests.post(
    "https://queue.modelrunner.run/luma/uni-1-max",
    headers=headers,
    json={
      "style": "auto",
      "prompt": "a lone lighthouse on a rocky cliff at golden hour, dramatic clouds, cinematic wide shot",
      "aspect_ratio": "16:9",
      "output_format": "jpeg",
      "enable_web_search": false,
      "reference_image_urls": null
    },
).json()

# Poll submitted["status_url"] until "COMPLETED", then:
result = requests.get(submitted["response_url"], headers=headers).json()

Input parameters

NameTypeRequiredDescription
promptstringyesThe text prompt describing the image to generate.
aspect_ratioenumnoThe aspect ratio of the generated image. Default: null.
styleenumnoThe visual style of the generated image. Default: "auto".
output_formatenumnoThe output image format. Default: null.
enable_web_searchbooleannoIf true, the model may consult the web while generating. Default: false.
reference_image_urlsarraynoOptional list of reference image URLs to steer composition (up to 9). Default: null.

Machine-readable: OpenAPI schema · llms.txt

Use Luma Uni-1 Max from Claude & Cursor (MCP)

Point Claude Code, Claude Desktop, Cursor, or any MCP client at the ModelRunner MCP server and Luma Uni-1 Max becomes a tool your assistant can call directly — it authorizes via OAuth (no API key in config) and runs this model with the run_model tool using the endpoint luma/uni-1-max.

MCP client config (Claude Desktop, Cursor)

{
  "mcpServers": {
    "modelrunner": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://mcp.modelrunner.run/mcp"]
    }
  }
}

Claude Code

claude mcp add --transport http modelrunner https://mcp.modelrunner.run/mcp

Then ask your assistant, for example: “Run luma/uni-1-max on ModelRunner to generate image”. MCP setup guide.

Model Details

Model Details

Luma Uni-1 Max turns a text prompt into a single high-fidelity image, the top tier of the Uni-1 family. It delivers richer detail and stronger prompt adherence than the standard Uni-1 tier, making it the pick when the output is a hero image rather than a quick draft. Beyond the prompt, you can switch to a manga visual style, let the model consult the web while generating, and pass up to nine reference images to steer composition.

## Best for - Hero-quality stills where detail and faithful prompt adherence matter more than speed - Rendering manga-style panels and illustrations via the dedicated manga style - Generating images that need accurate, up-to-date subject matter by grounding on web search - Steering composition, layout, or look from a set of reference images (up to nine) - Final marketing, editorial, or cover art from a detailed text description

## Choose another model when - You want a faster, lower-cost draft or to explore many variations in a batch — use the standard tier, luma/uni-1 - You need to edit or transform an existing image rather than generate one from a prompt — use an image-to-image or image-editing model - You need motion or video output — use a video model

## Tips - Write a detailed prompt; this tier rewards specificity in subject, lighting, and composition - Set `style` to `manga` for line-art/manga panels; leave it at `auto` for everything else - Turn on `enable_web_search` when the image depends on real, current subject matter the model might not know - Pass `reference_image_urls` (up to nine) when you want the result anchored to an existing look or layout

## Advanced Configuration - `style` selects the visual style: `auto` (default, model picks the rendering) or `manga` (manga/line-art panels). - `enable_web_search` (default `false`): when `true`, the model may consult the web for references while generating — useful for accuracy on real-world or current subjects, at the cost of some latency. - `reference_image_urls`: an optional list of image URLs (maximum nine) that guide composition and look. Supply public, downloadable URLs.

To run via the ModelRunner JavaScript client: ```js import { modelrunner } from "@modelrunner/client";

const result = await modelrunner.subscribe("luma/uni-1-max", { input: { prompt: "A stoic samurai overlooking a neon cyberpunk city at dusk", aspect_ratio: "16:9", style: "manga", enable_web_search: false, output_format: "png", }, }); ```