FLUX.2 Text to Image

FLUX.2 brings enterprise-grade efficiency and professional precision to text-to-image generation. It closes the gap between generated and real imagery with accurate hands, faces, and textures—all while respecting brand guidelines through hex-code color steering. [max] offers the highest quality with grounding search for real-time information.

Try it live — Test FLUX.2 [max], [pro], and [flex] in the playground. [klein] is available via API and on Hugging Face.

Photorealistic portrait generated by FLUX.2

Professional AI conference photograph with complex typography and hex color specifications

Capabilities

Explore why FLUX.2 is the choice for professional workflows.

Photorealism & Detail
Typography & Design
Grounding Search
Creative & Culture
Advanced Control

Close the gap between Generated and Real. FLUX.2 produces realistic image details up to 4MP: accurate hands, faces, fabrics, logos, and small objects that other models miss. Ideal for creative photography, e-commerce shots, and product marketing.

Structured PromptingUse structured prompts for precise control over generation. Perfect for production workflows and automation.

Example: Structured Prompting

{
  "subject": "Mona Lisa painting by Leonardo da Vinci",
  "background": "museum gallery wall, ornate gold frame",
  "lighting": "soft gallery lighting, warm spotlights",
  "style": "digital art, high contrast",
  "camera_angle": "eye level view",
  "composition": "centered, portrait orientation"
}

Exact Color ControlFLUX.2 supports precise color matching using hex color codes. Specify exact brand colors without approximation, making it ideal for professional design workflows.Method 1: Using Hex Codes in PromptsThe best way to include hex codes in your prompt is by signaling with keywords like “color” or “hex” followed by the code:Example: Apply hex colors to specific objects

Vase with gradient colors and flowers using hex color specifications

Integration

Integrate FLUX.2 text-to-image generation into your application in three steps.

Authenticate

Get your API key from the BFL Dashboard. You will use this in the x-key header.

Create a Generation Request

Send a POST request to the endpoint for your chosen model. The prompt is required.

[klein] 4B
[klein] 9B
[max]
[pro]

curl -X POST https://api.bfl.ai/v1/flux-2-klein-4b \
  -H 'accept: application/json' \
  -H "x-key: $BFL_API_KEY" \
  -H 'Content-Type: application/json' \
  -d '{
    "prompt": "A serene mountain landscape at golden hour, soft diffused light filtering through clouds, creating long shadows across the valley",
    "width": 1024,
    "height": 1024
  }'

curl -X POST https://api.bfl.ai/v1/flux-2-klein-9b \
  -H 'accept: application/json' \
  -H "x-key: $BFL_API_KEY" \
  -H 'Content-Type: application/json' \
  -d '{
    "prompt": "A serene mountain landscape at golden hour, soft diffused light filtering through clouds, creating long shadows across the valley",
    "width": 1024,
    "height": 1024
  }'

curl -X POST https://api.bfl.ai/v1/flux-2-max \
  -H 'accept: application/json' \
  -H "x-key: $BFL_API_KEY" \
  -H 'Content-Type: application/json' \
  -d '{
    "prompt": "Cinematic shot of a futuristic city at sunset, 85mm lens",
    "width": 1920,
    "height": 1080,
    "safety_tolerance": 2
  }'

flux-2-pro-preview reflects our latest advances and is continuously updated. Use flux-2-pro if you need a pinned model. See Preview Endpoint for details.

curl -X POST https://api.bfl.ai/v1/flux-2-pro-preview \
  -H 'accept: application/json' \
  -H "x-key: $BFL_API_KEY" \
  -H 'Content-Type: application/json' \
  -d '{
    "prompt": "Cinematic shot of a futuristic city at sunset, 85mm lens",
    "width": 1920,
    "height": 1080,
    "safety_tolerance": 2
  }'

Check Pricing

The API response includes pricing information. The cost field shows the credits charged for the request.

# Response includes cost information
{
  "id": "task-id-here",
  "polling_url": "<polling_url>",
  "cost": 3.0,           # Credits charged for this request
  "input_mp": 0.0,      # Input megapixels
  "output_mp": 2.07     # Output megapixels
}

Retrieve the Result

Poll the polling_url until the status is Ready.

# Poll until status is 'Ready'
curl -X GET "$POLLING_URL" \
  -H 'accept: application/json' \
  -H "x-key: $BFL_API_KEY"

Configuration

Parameter	Type	Default	Description
`prompt`	string	—	Required. Text description of the image to generate.
`width`	integer	`1024`	Output width in pixels. Must be a multiple of 16.
`height`	integer	`1024`	Output height in pixels. Must be a multiple of 16.
`seed`	integer	random	Set for reproducible results.
`safety_tolerance`	integer	`2`	Moderation level: `0` (strict) to `5` (permissive).
`output_format`	string	`"jpeg"`	Output format: `"jpeg"` or `"png"`.

[flex] only parameters:

Parameter	Type	Default	Description
`steps`	integer	`50`	[flex only] Number of inference steps. Maximum: 50, default: 50. Higher = more detail, slower.
`guidance`	float	`4.5`	[flex only] Guidance scale. Controls how closely the output follows the prompt. Minimum: 1.5, maximum: 10, default: 4.5. Higher = closer prompt adherence.

Resolution: Minimum 64x64, maximum 4MP (e.g., 2048x2048), recommended up to 2MP. Output dimensions are always multiples of 16.

Model Selection

	[klein] 4B	[klein] 9B	[max]	[pro]	[flex]
Best for	Real-time, high volume	Balance of speed & quality	Highest quality	Production at scale	Quality with control
Speed	Sub-second	Sub-second	< 15 seconds	< 10 seconds	Higher latency
Controls	Standard	Standard	Standard	Standard	Adjustable steps & guidance
Grounding search	No	No	Yes	No	No
Pricing	$0.014 +$ 0.001/MP	$0.015 +$ 0.002/MP	from $0.07/MP	from $0.03/MP	$0.06/MP

FLUX.2 [klein] 4B

Sub-Second GenerationFrom $0.014/image. Runs on consumer GPUs (~13GB VRAM). Apache 2.0 license. Perfect for real-time applications and local deployment.

FLUX.2 [klein] 9B

Frontier Quality at SpeedFrom $0.015/image. Matches models 5x its size. 9B flow model with 8B Qwen3 text embedder for superior prompt understanding. FLUX Non-Commercial License.

FLUX.2 [max]

Top-Tier QualityHighest image quality and prompt adherence. Includes grounding search for real-time information. Best for professional content needing the final touch.

FLUX.2 [pro]

Fast & EfficientBest balance of speed and cost. Ideal for high-volume applications needing fast turnaround.

FLUX.2 [flex]

Quality with ControlAdjustable steps and guidance. Best when you need fine-grained control over generation.

[klein] Prompting Guide

Narrative Prompting[klein] works best with prose-style prompts. Learn techniques for lighting, atmosphere, and scene description.

Get Started

Account Management

FLUX.2

FLUX.1 Kontext

FLUX1.1 [pro] Models

FLUX.1 Tools

API Integration

Capabilities

Integration

Configuration

Model Selection

FLUX.2 [klein] 4B

FLUX.2 [klein] 9B

FLUX.2 [max]

FLUX.2 [pro]

FLUX.2 [flex]

[klein] Prompting Guide

Get Started

Account Management

FLUX.2

FLUX.1 Kontext

FLUX1.1 [pro] Models

FLUX.1 Tools

API Integration

​Capabilities

​Integration

​Configuration

​Model Selection

FLUX.2 [klein] 4B

FLUX.2 [klein] 9B

FLUX.2 [max]

FLUX.2 [pro]

FLUX.2 [flex]

[klein] Prompting Guide

Capabilities

Integration

Configuration

Model Selection