FLUX.2 Klein 9B
Neural Network
FLUX.2 Klein 9B is an image generation model by Black Forest Labs, distilled in four steps for fast visuals.
Specifications
- Input
- text, images
- Output
- images
- Resolution
- 0.25 MP, 0.5 MP, 1 MP, 2 MP, 4 MP
- Aspect ratio
- 1:1, 16:9, 9:16, 3:2, 2:3, 4:3, 3:4, 5:4, 4:5, 21:9, 9:21
- File formats
- WEBP, JPG, PNG
Prompt cost
(per 1M tokens)
Video generation
(duration 8 sec., 720p)
Video generation
(duration 8 sec., 1080p)
How it works FLUX.2 Klein 9B?
FLUX.2 Klein 9B is a distilled version of FLUX.2 [klein] from Black Forest Labs that requires only four steps to generate an image from a text description or reference. The 9-billion parameter base architecture is designed for flexible customization and control, making it suitable for both quick draft iterations and final rendering. In BotHub, it is accessible without a VPN or foreign card: pay with a Russian card in rubles only for tokens actually used, which do not expire. Over 250 other neural networks are available in the same window, making it easy to compare results, and you can connect generation to your service via a unified OpenAI-compatible API without rewriting integrations. Requests are encrypted, and for companies, we offer contracts, invoices, and an admin panel with limits. FLUX Klein is useful for designers creating concepts and mood boards, marketers for ad and social media visuals, product teams for rapid prototyping, authors for covers and illustrations, and developers for streaming asset generation via API.Frequently asked questions about FLUX.2 Klein 9B
You can use generated results for commercial purposes. You own all rights to the content you create. The only restriction: make sure your prompt does not include copyrighted third-party material. You are responsible for respecting the rights to any input data.
This is a generative image model from Flux: it creates images from text descriptions and processes existing images based on references. It is suitable for illustrations, concepts, visuals for social media and advertising, covers, variations, and editing existing frames to fit a specific style.
You get a raster image ready for download and use in layouts. Specific resolutions and formats depend on request parameters, so they are not fixed in our card. The output limit is 4096 tokens, and the prompt and reference context is 4095 tokens.
A separate step-by-step reasoning mode is not noted in our data: this is a visual model that converts a description into an image, not a text response. To get a complex composition, describe the scene, lighting, angle, and style in more detail directly in the prompt or add a reference.
Function calling and structured JSON output are not noted in our data — the model's task is to return an image. However, you can access it via the unified OpenAI-compatible BotHub API and switch to other models in your code with one line without rewriting the integration.
Images — yes, image-to-image mode is confirmed: you pass the source frame as a reference and ask to change the style, details, or composition. Audio and video input are not noted in our data, so for such formats, it is more convenient to choose a specialized model in the same BotHub interface.
We do not have data on prompt languages, so we cannot claim anything — please check with your own requests. What is certain: the BotHub interface and Telegram bot work from Russia, without a VPN or foreign card, and payment is made with Russian cards in rubles based on actual usage.