FLUX.2 Pro Text to Image API

blackforestlabs/flux-2-pro/text-to-image
1K/2K · 8 sizes

FLUX.2 Pro Text to Image transforms natural language prompts into photorealistic visuals with Hex color control and native typography. It adheres to complex multi-subject layouts while rendering lifelike facial features.

Get API Key
Input
Required0/5000
Required
Required
OutputIdle

Generated images will appear here

Estimated cost: —
Continue using

Examples

pro-text-to-image-02-output.jpg

An interior photograph of a traditional Mongolian ger on a cold clear morning, viewed from just inside the low wooden door toward the back wall. Pale sunlight enters through the circular roof crown and forms a bright shaft through faint stove smoke. Exact colour palette: the painted roof crown ring and the radiating roof poles are lacquered in #C8102E crimson with fine #F2A900 saffron-yellow geometric knot patterns; the two central support columns are painted #1B4D89 deep blue; the felt walls are undyed #EDE3D1 wool cream; the floor carpet is #7A2E1F madder brown with cream borders. A cast-iron stove with a steaming kettle stands in the centre, a low painted table holds two small bowls of milk tea, and a folded deel robe lies on a wooden bed at the left. Every coloured surface matches the specified hex values under natural daylight, with authentic hand-painted brush texture, wool fibre detail and soft bounce light. Documentary interior photography, 24mm lens, no people, no text, no logos, no watermark.

pro-text-to-image-03-output.jpg

A wide-angle industrial documentary photograph deep underground in a newly excavated metro tunnel. The enormous circular cutterhead of a tunnel boring machine, about twelve metres across, fills the right half of the frame, its steel disc cutters worn and caked with damp grey clay. Three workers in orange high-visibility overalls, white helmets with headlamps and mud-streaked boots inspect the cutters from a narrow steel gantry at mid-height: one points a flashlight at a chipped cutter, one takes notes on a clipboard, and the third holds a long torque wrench. Along the left, curved precast concrete tunnel rings recede into the distance, lit by a string of cool-white work lights that create repeating pools of light and hard shadows. Water drips from the tunnel crown and glistens on the rails of a small muck-removal train. Strong sense of scale between people and machine, realistic steel, concrete and clay textures, humid haze catching the light beams, truthful colour. No brand names, no lettering on the machine or clothing, no watermark.

FLUX.2 Pro Text to Image

FLUX.2 Pro Text to Image is a 32-billion-parameter foundation model engineered by Black Forest Labs. Integrating a Mistral-3 visual language model with rectified flow matching, it delivers photorealistic fidelity, Hex color precision, and native English typography. The model outputs up to 2K resolution, starting at 6 credits ($0.030) per generation.

Why Choose FLUX.2 Pro Text to Image?

  • Exact Hex-Code Color ControlIncorporate precise hexadecimal color codes directly into your prompts to match brand palettes without approximation or drift.

  • Native In-Image TypographyRenders legible, sharp text, signs, and labels directly on canvas by enclosing required phrases in quotation marks.

  • 32B Flow-Matching & Mistral-3 VLMCombines a 32-billion-parameter rectified flow backbone with a 24B vision-language model for deep world knowledge and spatial reasoning.

  • Native 2K Output Across 8 Aspect RatiosDelivers up to 4MP output across 1:1, 4:3, 3:4, 16:9, 9:16, 3:2, 2:3, and auto dimensions with crisp micro-textures.

  • Predictable Pricing & Automated RefundsCharges 6 credits for 1K and 9 credits for 2K with full automatic refunds if a generation task fails.

Parameters

ParameterRequirementDescription
promptRequired

Detailed description of the target image, including subjects, lighting, perspective, and enclosed quoted typography. Length: 3–5,000 characters.

sizeRequired

Canvas aspect ratio preset. Accepted values: 1:1, 4:3, 3:4, 16:9, 9:16, 3:2, 2:3, auto.

1:14:33:416:99:163:22:3auto
resolutionRequired

Output rendering resolution. Accepted values: 1K (standard HD) or 2K (ultra-high definition up to 4MP).

1K2K

How to Use

  1. Formulate Your Prompt and Color CodesSpecify focal subjects, lighting mood, and composition. Add hex color codes for brand consistency and enclose readable text in double quotation marks.

  2. Select Aspect Ratio and ResolutionChoose an aspect ratio such as 16:9 or 1:1, and select 1K for standard output or 2K for ultra-fine 4MP rendering.

  3. Submit Generation RequestSend a POST request with your API key to the unified submit endpoint and store the returned task_id.

  4. Poll Status and Retrieve OutputQuery the status endpoint every 2 to 5 seconds until status reaches finished, then download your rendered image URL.

  5. Integrate Production WebhooksOptionally configure a top-level callback_url in automated pipelines to receive instant POST notifications upon task completion.

Pricing

1K generation costs 6 credits ($0.030) and 2K generation costs 9 credits ($0.045) per image with zero monthly commitments.

UsageRateDetails
1K Standard Generation6 credits/generation · $0.030Applies to 1K resolution across all eight aspect ratio presets
2K Ultra-HD Generation9 credits/generation · $0.045Applies to 2K resolution (up to 4MP) across all eight aspect ratio presets

Best Use Cases

  • Brand Marketing & AdvertisingGenerate campaign visuals with exact hex color adherence and crisp promotional text for digital ads and print billboards.

  • E-Commerce Product ScenesProduce photorealistic commercial lifestyle scenes with physically accurate lighting, reflections, and natural material textures.

  • Editorial & Graphic PostersDesign high-resolution posters and infographics featuring clean, legible headline typography embedded directly into the artwork.

  • UI Mockups & Concept VisualizationDraft app interfaces, packaging concepts, and storyboards with balanced spatial composition and natural depth of field.

Pro Tips

  • Use Quotation Marks for In-Image Text:Enclose any required letters or slogans in double quotes (e.g., a sign reading "NORDIC ROAST") to guide accurate typography.
  • Leverage Exact Hex Color Codes:State exact hex values like #FF5722 or #009688 to align visual tones strictly with corporate brand guidelines.
  • Define Spatial Composition and Lighting:Specify camera angles, depth of field, and lighting direction (e.g., dramatic side rim light) to enhance spatial realism.
  • Choose 2K for Commercial Print and Zoom:Select 2K resolution when preparing assets for large displays or high-density prints requiring fine micro-textures.
  • Set Up Callbacks for High-Volume Systems:Provide a callback_url in backend systems to process image completions asynchronously without sustained polling loops.

Notes

  • Prompt Length Boundaries:The prompt parameter requires between 3 and 5,000 characters, including at least one non-whitespace character.
  • Pure Text-to-Image Scope:This endpoint accepts text prompts only. For multi-image reference workflows, use FLUX.2 Pro Edit.
  • Asynchronous Task Lifecycle:Tasks execute asynchronously. Status moves from not_started and running to terminal states finished or failed.
  • Credit Refund Guarantee:Credits deducted at submission are automatically and fully refunded if a generation task encounters a processing error.

Related Models

FLUX.2 Pro Text to Image API Frequently Asked Questions

What is the FLUX.2 Pro Text to Image API?

FLUX.2 Pro Text to Image is a Black Forest Labs model for enterprise-grade text-to-image generation. It transforms natural language prompts into photorealistic images up to 2K resolution with exact hex-code color steering, complex in-image typography, and anatomically precise human details. Built on a 32B flow-matching transformer coupled with a 24B Mistral-3 vision-language model and custom VAE, it preserves physical spatial balance and brand color standards while synthesizing intricate textures. You can call it programmatically or try it from the playground above.

How do I specify exact brand colors in FLUX.2 Pro Text to Image?

You can insert standard hexadecimal color codes directly within your prompt text, such as "a modern chair in brand color #0F52BA on a polished concrete floor". The model interprets hex codes natively to guide color saturation and hue without perceptual approximation.

Can FLUX.2 Pro Text to Image render legible text within generated images?

Yes. Enclosing desired words or sentences in double quotation marks inside your prompt enables the model to render crisp, readable typography on signs, packaging, posters, and user interface mockups without typical letter distortion.

What is the difference between 1K and 2K resolution in FLUX.2 Pro Text to Image?

1K resolution produces standard high-definition visuals at 6 credits ($0.030), ideal for rapid concepting and digital marketing. 2K resolution delivers up to 4-megapixel (4MP) imagery at 9 credits ($0.045), providing superior micro-textures, sharper typography, and print-ready fidelity.

Which aspect ratio presets are available for FLUX.2 Pro Text to Image?

The size parameter accepts eight presets: 1:1, 4:3, 3:4, 16:9, 9:16, 3:2, 2:3, and auto. You can choose vertical ratios for mobile stories, horizontal ratios for desktop banners, or square ratios for social posts.

How does credit billing and refund protection work for FLUX.2 Pro Text to Image?

Each generation deducts 6 credits for 1K or 9 credits for 2K upon request submission. If a task fails to complete or encounters an unexpected processing timeout, the system automatically returns all deducted credits to your balance.

When should I choose FLUX.2 Pro Text to Image over FLUX.2 Flex Text to Image?

Choose FLUX.2 Pro Text to Image for commercial production workflows that demand high visual fidelity, rapid generation turnaround, and budget-friendly pricing at 6 credits. Choose FLUX.2 Flex Text to Image when your project requires fine-grained control over specialized typography layouts and delicate artistic styling.