How to use Twilix
1. Selecting and Ingesting Data
To start using our API endpoints, you need to pass in data so we have enough context to give you customized outputs. Depending on the endpoint you’re using, the data you pass in can either be from within your own vectorstore, or from an external data source.
For example, if you want to build a chatbot using our generative QA endpoint, you need specify what collection we should use to retrieve the relevant imformation.
Depending on your usecase, you might want to use your own vectorestore. We support a wide range of vectorstores, please contact us for custom vectorestore integration. Using external data will not affect the quality of results.
To start data ingesting, use our endpoints under the Ingest Data section.
2. Start using the AI services endpoints
Once you have defined the data you want to use in the request body, you can now run queries on them.
We offer various endpoints for communicating with your data. Take a look under Query Data sections for how to use each endpoint.
Twilix is powered by BlitzChain under the hood. BlitzChain is our in-house version of LangChain that is superior in accuracy, speed, and cost. You don’t directly interact with BlitzChain, but you can influence what it outputs by defining what data to feed it in the request body.