What's New

We're thrilled to update you on some of the newest enhancements and give you a sneak peek into what’s coming next.

🦜🔗 ​LangChain​ Integration

Large language models (LLMs) are emerging as a transformative technology, enabling developers to build applications that they previously could not. However, using these LLMs in isolation is often insufficient for creating a truly powerful app - the real power comes when you can combine them with other sources of computation or knowledge.

By embedding the SearchAPI, the Large Language Model can respond by querying real-time data to answer questions about current events. Its usage is particularly interesting with the "Self-ask with search" agent. This agent is equivalent to the original ​self-ask with search paper​, where a Google search API was provided as the tool.

Another common user case is summarizing long documents that naturally run into context window limitations. Here is an example ​using SearchApi to load a lengthy YouTube video transcript​, then summarize and distill it with a refine chain.

SearchApi is now natively available as part of ​langchainjs​ and ​langchain​ libraries.

For a quick overview, check out this tweet on SearchApi integration in LangChainJS.

Here is the official SearchApi documentation on its usage in Python and Javascript:
  •  Python examples​
  •  JavaScript/TypeScript examples​​

🌟 Freshly Released Engines:

  ✔️  Google Lens API​ - shipped in less than 24 hours based on ​@levelsio tweet on Twitter​. Useful as a tool in ​affiliate marketing​ for fetching product sources, links, prices, and stock information.
  ✔️  YouTube Transcripts API​
  ✔️  Amazon Search API​

🔧 Engine Improvements:

Google
  •  Added the ability to select a custom date range.
  •  Fixed `inline_videos` parsing on mobile
  •  Added `source` to desktop & mobile organic results
  •  Fixed `detected_location` in `search_information`
  •  Fixed `date` and `snippet` identification in organic results

⏭️ What's coming next?

  •  Google Maps engine
  •  Amazon Product engine
  •  Integration examples in multiple languages and libraries
  •  LangChain integration (Ruby)

Keep an eye out for further updates! We're dedicated to ensuring you have the best experience.