Creating an assistant

After login into your dashboard click on the "Create New Assistant Button" this will open up the sources selection.

We currently support multiple types of information/knowledge sources meaning that you can feed your assistant with information from your business in the formats of PDF, Docx, CSV , TXT, MP3, MP4 and or you can simply add the link of your website to make the assistant have access to the information available on your site.

Then select the AI and Embedding model you want for your assistant, we currently support a variety of advanced language models and embedding models. This includes OpenAI's GPT-3.5-turbo, GPT-3.5-turbo-16K, GPT-4, GPT-4-0613, and GPT-3.5-turbo-instruct. We also support Anthropic's Claude-1, Claude-instant-1, and Claude-2, as well as Google's PaLM 2 (chat-bison-001).

What are embeddings? Embeddings in the context of machine learning are a way to convert complex data, like text or images, into a format that a computer can understand and work with.

Think of it like translating a human language into a machine language. For example, when dealing with text, the computer doesn't understand words or sentences like humans do. So, we use embeddings to translate words or sentences into numerical vectors (a list of numbers) that represent the meaning and context of the words.

This allows the computer to "understand" and process the information in a way that makes sense to it, and it can then use this information to do tasks like language translation, recommendation systems, or image recognition.