Krea 2 Medium
Neural Network
Krea 2 Medium is an image generation model from Krea that provides stable results based on text descriptions.
Specifications
- Input
- text, images
- Output
- images
- Resolution
- 1K, 2K, 4K
- Aspect ratio
- 1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 16:9, 9:16, 21:9
How it works Krea 2 Medium?
Krea 2 Medium is a balanced image generation model from the Krea team, designed as an economical entry point for visual tasks. Extensive post-training provides predictable results: frames are stable and consistent, making the model suitable for both one-off experiments and working on image series. In BotHub, it is accessible without a VPN or foreign card, and you pay with a Russian card in rubles, only for what you generate. Over 250 models are available in one window: you can compare visual styles and switch between them, while a unified OpenAI-compatible API allows you to integrate generation into your product without rewriting integrations. Data is transmitted in encrypted form, and companies have access to contracts, invoices, and an admin panel with limits. It is useful for covers and illustrations for articles and social media, visuals for ad creatives and landing pages, concept art and mood boards at the start of a project, product cards and backgrounds, and for quickly testing dozens of ideas before drawing the final layout.Frequently asked questions about Krea 2 Medium
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.
krea-2-medium creates images: it generates a picture from a text description and modifies an existing one if you upload it as a reference. It is suitable for concepts, illustrations, visuals for social media and advertising, quick variations of an idea, and refining existing frames.
You get a finished image as output, not text. Exact resolution parameters and file formats depend on request settings and are specified in the technical section of the page. For a detailed result, describe the scene, lighting, angle, and style in detail — the model relies on the prompt formulation.
A separate step-by-step reasoning mode is not noted in our data: this is a generative image model, it renders the result immediately upon request. If the idea is complex, break it down yourself — describe the composition and details in order, or refine the image through iterations in image-to-image mode.
Function calling and structured JSON output are not noted in our data: the model's result is an image. However, you can access it via the unified OpenAI-compatible BotHub API, so integrating generation into your service and application logic is easy.
Images — yes: krea-2-medium accepts an image as input and works in image-to-image mode when you need to change the style, refine a frame, or create variations based on a reference. Audio and video are not supported as input formats for this model, so rely on text and images.
We have no data on query language, so formulate the prompt as you prefer, check the result, and refine the scene description if necessary. BotHub itself is fully Russian-language: interface, payment with Russian cards in rubles, and access from the country without a VPN.