How it works Recraft V4 Vector?
Recraft V4 Vector is the vector version of the Recraft V4 model by Recraft: it accepts text and images as input and outputs results as SVG vector graphics in various aspect ratios. Unlike raster versions, the result consists of curves and fills, allowing for scaling without quality loss and further editing. In BotHub, the model is accessible from Russia without a VPN or foreign card; payment is in rubles and based on actual usage, with no expiring tokens. Over 250 other neural networks are available in the same window, allowing you to combine vector, raster, and text models in one project or call them via a unified API. Designers create logos and brand marks, marketers generate icons and illustrations for landing pages and presentations, and product teams build icon sets for interfaces. For printing, you get a file that looks equally clean on a business card or a banner. You can start with a short text description or upload a reference to get its vector interpretation.Frequently asked questions about Recraft V4 Vector
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 Recraft model for generating graphics from text descriptions and references: logos, icons, pictograms, illustrations, and interface elements. It is suitable for designers, marketers, and developers when a clean vector image is needed that can later be scaled for banners, printing, or favicons.
The model specializes in vector graphics: clean shapes and contours that scale without pixelation or blurry edges, unlike standard raster images. Specific resolutions and file formats are not listed in our data, so please rely on the results you receive in the BotHub interface.
A separate reasoning mode before answering is not noted in our data, so we cannot confirm it. For graphic tasks, this is usually unnecessary: the result depends more on the prompt, so describe the object, style, palette, and composition in detail.
Function calling and strict structured JSON output are not noted for this model in our data. If such behavior is critical for your service, connect to the unified OpenAI-compatible BotHub API and select a model with confirmed function calling — you won't have to rewrite the integration.
The model accepts images and documents as input: you can send a reference and get a variation or description based on it. Audio and video support is not noted in our data. Context is 65,536 tokens, maximum per response is 58,982 tokens.
We have no confirmed data on prompt and output languages, so we cannot make any claims — it is easier to test it with your own task. On the service side, everything is in Russian: the BotHub interface and support, payment with Russian cards in rubles, and access from the Russian Federation without a VPN or foreign card.