Ximilar runs pre-built computer vision services accessible through a no-code app or a developer API, covering fashion and home decor tagging and search, AI recognition and grading for collectibles such as trading cards, stock photo tagging and similarity search, reverse image search, background removal, image upscaling, and OCR, including an OCR-plus-ChatGPT combination for extracting and interpreting text from images. Beyond the pre-built services, its Computer Vision Platform lets a team train custom image classification, object detection, instance segmentation and person detection models on its own annotation tool, and combine multiple models into a single processing flow. Founded in Brno, Czech Republic in 2016 by David Novak, Michal Lukac and Vitezslav Valka, the company prices access through monthly API credits rather than a flat subscription.
Because pricing is credit-based and different operations consume credits at different rates depending on the specific service used, a team should run its expected image volumes through Ximilar’s own credit supply calculator before committing to a tier, rather than assuming a plan’s headline credit number translates directly into a fixed number of processed images. As with any pre-trained recognition or tagging service, accuracy on a highly specialized or unusual visual category may benefit from the custom model training option rather than relying on the general-purpose pre-built services alone.
Free is €0 forever for 1,000 API credits a month, no credit card required, with limited email support, suited to individuals or teams starting with no-code computer vision. Business is €59 (about $69) a month for 100,000 API credits, standard support and one consultation with an ML expert, aimed at businesses automating image processing. Professional is €499 (about $579) a month for 1,000,000 API credits, priority support and regular ML expert consultations, aimed at companies processing large data volumes. Additional credit packs can be purchased on any tier, and custom on-premise or edge-device deployments are available by contacting sales.
