Your LLMs deserve better input.
Readers can do two things:
It converts any URL to https://r.jina.ai/https://your.url LLM friendly input. Increase the output of agents and RAG systems for free.
it uses https://s.jina.ai/your+query Search the network for a given query. This allows your LLMs to gain the latest world knowledge from the web.
Live demonstration:
Or just visit these websites https://r.jina.ai/https://github.com/jina-ai/reader、https://s.jina.ai/Who%20will%20win%202024%20US%20presidential %20election%3F Look at yourself.
Feel free to use the Reader API in production. It is free, stable and scalable. We are actively maintaining it as one of the core products of Gena’s artificial intelligence
Use r.jina.ai to obtain a single URL
Just add it in front of any URL https://r.jina.ai/ That’s enough. For example, to set the URL https://en.wikipedia.org/wiki/Artificial_intelligence To convert to LLM-friendly input, use the following URL:
https://r.jina.ai/https://en.wikipedia.org/wiki/Artificial_intelligence
All images on this page that lack the alt tag will be automatically titled by VLM (Visual Language Model) and formatted as! (Image [idx]: [VLM_caption])[img_URL] 。This should provide your downstream plain text LLM with enough hints to include these images in reasoning, selection, and summary.
Use r.jina.ai for complete website crawling (Google Colab)
Use s.jina.ai for web search
simply https://s.jina.ai/ Add it to your search query. Please note that if you use it in code, make sure you encode your search query first, for example if your query is Who will win 2024 US presidential election? Then your website should look like this:
https://s.jina.ai/Who%20will%20win%202024%20US%20presidential%20election%3F
Behind the scenes, Reader searches the web, gets the top 5 results, visits each URL, and applies r.jina.ai to it. This is different from many web search function-calling in the proxy/RAG framework, which typically only returns the title, URL, and description provided by the search engine API. If you want to read a result more in-depth, you must get the content from that URL yourself. Using Reader, http://s.jina.ai Automatically retrieves content from the first 5 search result URLs for you (reuse http://r.jina.ai The technology stack behind it). This means you don’t have to deal with browser rendering, blocking, or any JavaScript and CSS-related issues yourself.
- jina reader (open source): https://github.com/jina-ai/reader/
- https://markdowndown.vercel.app
- Web-scraper(open source): https://github.com/zzzgydi/webscraper
- code-html-to-markdown: https://github.com/SivilTaram/code-html-to-markdown (Preference for better handling of code blocks)
If you want to learn more, you can click on the link below the video.
Thank you for watching this video. If you like it, please subscribe and like it. thank
Oil tubing: