FLUX.2-klein-4B

Summary: FLUX.2-klein-4B is a compact, efficient text-to-image model developed by Black Forest Labs, designed to deliver high-quality image generation with a reduced 4B parameter footprint. Building on the FLUX architecture, the klein variant prioritises speed and resource efficiency while maintaining strong visual fidelity, making it well-suited for content creation, rapid prototyping, and applications where low-latency generation is essential.

Image quality	Speed	Input	Output
Medium	High	Text	Image

Central parameters

Description: Compact 4B-parameter image generation model optimised for fast, high-quality visual output with low resource overhead.

Model identifier: black-forest-labs/FLUX.2-klein-4B

IONOS CLOUD AI Model Hub lifecycle and alternatives

IONOS CLOUD start date	End of Life	Alternative	Successor
June 16, 2026	N/A	N/A

Origin

Provider	Country	License	Flavor	Release
Black Forest Labs	Germany	License	Base	TBD

Technology

Input Length	Iterations	Quantization	Multilingual	Further details
256	4	8bit	Yes	Hugging Face

Modalities

Text	Image	Audio
Input	Output	Not supported

Endpoints

Chat Completions	Embeddings	Image generation	Image editing
Not supported	Not supported	v1/images/generations	v1/images/edits

Features

Streaming	Reasoning	Tool calling	Image editing
Not supported	Not supported	Not supported	Supported

Usage example

Parameter	Values
n	An integer between 1 and 10 (inclusive). Limit: The total output cannot exceed 12,582,912 pixels per request (n × width × height). For example, if requesting 2048x2048 images, n can be at most 3.
output_format	jpg, jpeg, png, webp
size	A string in widthxheight format. Both dimensions must be between 64 and 2048 (inclusive) and are automatically rounded down to the nearest multiple of 16. The default for image generation is 1024x1024. For image editing, the default is the input image's aspect ratio and size, scaled down to fit these limits. Examples: 1024x1024, 1536x1024, 1024x1536, 2048x2048, 2048x1152.

Image editing requires a single input image with a minimum dimension of 64 pixels on each side and an aspect ratio no more extreme than 8 to 1 (or 1 to 8). Requests that do not meet these limits are rejected with a 400 Bad Request error detailing the exceeded limit.

Generate an Image

The following example demonstrates how to generate an image using FLUX.2-klein-4B.

API Endpoint: POST https://openai.inference.de-txl.ionos.com/v1/images/generations

Request:

{
  "model": "black-forest-labs/FLUX.2-klein-4B",
  "prompt": "A futuristic city with flying cars at sunset",
  "n": 1,
  "size": "1024x1024",
  "output_format": "jpeg"
}

Response:

{
  "created": 1589478378,
  "data": [
    {
      "b64_json": "iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAQAAAC1HAwCAAAAC0lEQVR42mNk+A8AAQUBAScY42YAAAAASUVORK5CYII="
    }
  ]
}

Edit an image

The following example demonstrates how to edit an image using FLUX.2-klein-4B.

API Endpoint: POST https://openai.inference.de-txl.ionos.com/v1/images/edits

Request format: multipart/form-data

Request:

{
  "model": "black-forest-labs/FLUX.2-klein-4B",
  "prompt": "Add a rainbow in the sky above the cityscape",
  "url": "https://example.com/cityscape.png",
  "n": 1,
  "size": "1024x1024",
  "response_format": "b64_json"
}

Response:

{
  "created": 1589478378,
  "data": [
    {
      "b64_json": "iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAQAAAC1HAwCAAAAC0lEQVR42mNk+A8AAQUBAScY42YAAAAASUVORK5CYII="
    }
  ]
}

Rate limits

Rate limits ensure fair usage and reliable access to the AI Model Hub. The default rate limits apply to this model, separately from other models.

The default limit is 5 requests per second, with bursts of up to 10 requests. Each request counts once, however many images it generates.