Skip to content

FLUX1.1 [pro] Ultra

bfl/flux-pro-1.1-ultra

FLUX1.1 [pro] Ultra is Black Forest Labs's highest-resolution tier of the FLUX 1.1 Pro model. It adds raw image format support and expanded resolution capabilities for print, large-format, and photorealistic output use cases.

Image Gen
index.ts
import { experimental_generateImage as generateImage } from 'ai';
const result = await generateImage({
model: 'bfl/flux-pro-1.1-ultra',
prompt: 'A red balloon on a wooden table.'
});

What To Consider When Choosing a Provider

  • Zero Data Retention

    AI Gateway does not currently support Zero Data Retention for this model. See the documentation for models that support ZDR.

    Authentication

    AI Gateway authenticates requests using an API key or OIDC token. You do not need to manage provider credentials directly.

FLUX1.1 [pro] Ultra's raw image format output is particularly valuable for post-production pipelines where generated images will be retouched, color-graded, or composited. Raw files retain more tonal information than compressed formats.

When to Use FLUX1.1 [pro] Ultra

Best For

  • Print and large-format output:

    Advertising materials, billboards, posters, and print publications that need maximum image resolution and high pixel density

  • Post-production pipelines:

    Raw format output retains full tonal information for professional retouching, color grading, or compositing

  • High-resolution stock imagery:

    Commercial generation where output quality and fine photorealistic detail are the primary requirement

  • FLUX.1 quality ceiling:

    Applications that justify the additional per-image cost for the highest visual quality in the FLUX.1 generation

Consider Alternatives When

  • Cost efficiency:

    FLUX 1.1 Pro delivers the same model at lower cost per image when standard resolution output is sufficient

  • FLUX.2 architecture features:

    FLUX.2 Pro or FLUX.2 Max provide multi-reference input, 4 MP editing, and improved world knowledge architecture

  • Real-time generation speed:

    Ultra's higher-quality generation isn't optimized for sub-second interactive throughput

Conclusion

FLUX1.1 [pro] Ultra extends FLUX 1.1 Pro to its maximum quality ceiling with raw image format output and higher resolution support. For professional creative pipelines where AI-generated images serve as source assets rather than final deliverables, Ultra provides the output quality and format characteristics that standard tiers don't.

FAQ

This page lists the current rates. Multiple providers can serve FLUX1.1 [pro] Ultra, so AI Gateway surfaces live pricing rather than a single fixed figure.

Raw image format output and expanded resolution capabilities. Standard FLUX 1.1 Pro produces compressed image outputs. Ultra retains more tonal information through raw format support and supports higher output dimensions.

Raw image files preserve the full tonal range of generated content without lossy compression artifacts. For post-production workflows involving retouching, color grading, or compositing, raw files give creative teams significantly more latitude than compressed formats like JPEG.

FLUX.2 Max is part of the FLUX.2 generation with multi-reference input, 4 MP resolution, and improved world knowledge architecture. Ultra is the FLUX.1 generation's highest quality tier. If your workflow specifically benefits from raw format output in the FLUX.1 lineage, Ultra is appropriate. If you need multi-reference and world knowledge capabilities, FLUX.2 Max is more suitable.

Higher-resolution generation generally requires more compute per image. Ultra prioritizes maximum quality and resolution over generation speed, so it isn't the right choice for latency-sensitive or real-time workflows.

FLUX1.1 [pro] Ultra is a text-to-image model. For editing existing images, the Kontext models (Pro, Max) and Fill Pro are purpose-built for editing and inpainting workflows respectively.

No. FLUX1.1 [pro] Ultra is a dedicated image generation model with no conversational context or chat interface. Use generateImage from the AI SDK to send requests.