Did you know you can make Tess AI's chat even more powerful and personalized? By adding a file, you can give specific context to your conversations, making the answers way more relevant and accurate. In this guide, we'll show you how to do that in a simple and effective way.
Tess AI supports several file formats:
Audio
Google Sheets
PPT
Web pages
Excel
CSV
Others
Adding a file is a simple process:
Open the Tess AI chat;
You can drag the file you want into the command chat area (up to five different files at once);
Or you can find the plus (+) icon - the first icon to the left side of the command field;
Click the icon to open the options;
Select “Add photos & files” or “Open photos & files”;
Do the file upload you want;
When adding files to your knowledge base, keep in mind:
The maximum file size is 200Mb
You can add files until you reach the 200k Token limit
Keeping your files used as knowledge base organized is key for Tess AI chat to work well. To remove files you don't need anymore:
Organize your files: Use descriptive names to make them easier to find
Stay within the limits: Avoid going over 200MB per file
Mix different file types: Use PDFs for long documents, audio files for transcripts, and web pages for dynamic info
In this case, we recommend using the RAG methodology, a feature that lets you expand your chat agents' knowledge beyond what the models support in content volume, unlocking possibilities for creating increasingly well-trained chats.
LLMs are trained on massive datasets, which allows them to generate text, translate languages, and answer questions impressively. But this knowledge is static up to a point—so, if the LLM has a context window of 200k tokens, the content you use to train it as a knowledge base can't be bigger than that.
RAG breaks that barrier, letting your chats access and read extra info besides the Knowledge Base attachments, adding to the context and bringing more complete responses.
Basically, RAG reads by making connections and mentioning certain topics. For example, you can add a 50-page file, but in RAG, the AI will only look for what's related to your question. Since it's optimized like that, RAG allows an increase in the context window for that LLM.
When you add a knowledge base, there's a small processing fee that will only use your credits once, just to access your loaded base.
And it's usually very little, so don't worry. If you don't add an extra base and use the unlimited model, you won't spend any of your credits!
When you add a file as a knowledge base to Tess AI's chat, you're really boosting what it can do. Try uploading different types of files and see how it can change your interactions with the AI.
Remember: the quality of your knowledge base has a direct impact on the quality of Tess AI's answers. Put some time into making a solid and relevant base, and you'll see awesome results in your conversations.
