Krea K3 Edit
Neural Network
Krea K3 Edit is a Krea model for editing images using references and text descriptions of edits.
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 K3 Edit?
Krea K3 Edit is an image editing model from Krea: you provide one to four references and a text description of the edit, and receive one image as output. The model works based on instructions for an existing image rather than generating a frame from scratch, and relying on multiple references helps maintain consistency of characters, objects, or styles between edits. In BotHub, the model is accessible from Russia without a VPN or foreign card; payment is made with Russian cards in rubles on a pay-as-you-go basis, with no subscription required. Alongside it, in one window, are over 250 other neural networks: you can create a reference in one model and apply edits in K3 Edit without changing services. For product tasks, there is a unified OpenAI-compatible API, and for teams, there is a contract, invoice, and admin panel with limits. It is useful for designers iterating on layouts and banners, marketers adapting visuals for different campaign formats, retouchers for background replacement and detail cleanup, shops for consistent product cards, and content creators for series of frames with a single character.Frequently asked questions about Krea K3 Edit
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.
k3-edit from Krea works with images: it creates them from text descriptions and processes existing images based on references. It is suitable for composition edits, replacing details, creating variations of a single frame, and preparing visuals for banners, product cards, and social media.
You get an image as output, not text. The specific resolution and format depend on your request and generation parameters; exact values are indicated in the technical header of the page. For edits by reference, the model relies on the uploaded image and preserves its key details.
A separate reasoning mode before answering is not noted in our data — this is an image model, and all work is directed towards the image, not text-based reasoning. The quality of the result here depends more on how detailed your description of the desired frame is.
In our data, function calling and strict JSON output are not noted for k3-edit: the model's task is to output an image. However, you can access it via the unified OpenAI-compatible BotHub API, so integrating generation into your service is easy.
Images — yes: k3-edit accepts an image as input and processes it according to your description; this is image-to-image mode. There are no notes in our data regarding audio or video input, so for such files, it is better to choose a specialized model in BotHub.
We have no data on supported prompt languages, so we will not make any promises here. Practical advice: describe the frame specifically — object, style, light, angle. BotHub itself is Russian-speaking: interface, support, and payment with Russian cards, without VPN or foreign cards.