Built-in Interfaces

These docs are outdated! Please check out https://docs.titanml.co/docs/category/titan-takeoff for the latest information on the Titan Takeoff server. If there's anything that's not covered there, please contact us on our discord.

The Takeoff server was developed specifically for speed of development - to shorted the time between downloading a model and being able to interact with it and test it.

To that end, we created two user interfaces that automatically ship with the takeoff server that make it as easy as possible to interact with your models in a way that you might be familiar with.

There are two interfaces - Chat and Playground

Chat Interface

To use the chat interface navigate to localhost:8000/demos/chat/ in your browser with the Takeoff server running

The chat interface is designed to mimic the ChatGPT interface.

Your messages and the bot's responses are sent back and forth via the api each time you send a message, meaning TitanBot has a memory of the conversation up to that point.

Press Toggle Settings in the top right hand corner of the screen to change temperature and other generation settings.

This interface is good for testing out chat-optimized models like the llama 2 chat variant.

Playground Interface

To use the chat interface navigate to localhost:8000/demos/playground/ in your browser with the Takeoff server running

The playground interface is based on the OpenAI playground. This is a way to quickly test non-chat behaviour of models. Any text you add to the box is send to the API, and the model tries to complete it.

Use this to quickly test out question answer abilities, summarization abilities, or anything anything else.

There are two commands to remember:

Shift + Enter: New Line

Enter: Ask the model for a completion.

Again, use Toggle Settings in the top right corner to change generation parameters.