A couple of APIs are hosted by myself and are required for the app. They are free and I intend to continue running them as they’re basically free. However, you are welcome to host it yourself using the code provided. Both are hosted on GCP. No support is provided if you are trying to host these yourself though as they already work preconfigured with the template.
If you do not want to set these APIs up yourself, you can ignore this page.
This splits text into chunks. By default, it will split into chunks of n words. There is a fallback number of characters, so, for example, if it reaches 1250 characters and has still not split the chunk, it will split there even if it’s in the middle of a word.
This formats a JSON file that is used in the Pinecone upsert process. It essentially takes a list of text chunks and an array of an array vector values and formats it in a way that Pinecone can understand. It also generates a unique ID for each vector text/value pair.