Jina Search Foundation API
Read URLs and search web for better grounding LLMs.
Jina AI's Reader API is a powerful tool designed to optimize web content for Large Language Models (LLMs). It converts URLs, local HTML and PDF files into clean, structured text, removing extraneous elements and making it ideal for grounding, reasoning and retrieval-augmented generation (RAG) systems. This free API includes functionalities to directly read URLs, conduct web searches and perform fact-checking, by simply prepending specific prefixes to your requests. Notably, Reader API also handles image captioning and can extract data from embedded iframes and shadow DOMs. It offers advanced parameters, such as custom CSS selectors, browser viewport configuration and cookies forwarding.
The Reader API's unique selling point lies in its ability to provide not just the textual content but also additional structured information. It allows users to retrieve and include links, images, and captions as part of the output that aids LLM comprehension. With this API, users can reduce LLM hallucination, enhance factuality of LLM outputs and improve the integrity of generated content. This free service is built on a scalable infrastructure and offers a million free tokens with every API key, making it a reliable, accessible, and cost-effective solution for developers, researchers, and businesses alike.
