If you've become accustomed to the Vault but find yourself wanting more, consider exploring Stacks.
Stacks was designed as a tool within the platform for those with innovative ideas to construct the foundational AI framework for applications. It enables you to link together multiple AI models to produce various outputs, extending beyond simple query responses. We even refer to it as a benchmark tool for this very reason.
The image below shows the basic structure of the Stacks testing application
1. Save (this saves the stack state)
2. Add new models to this stack
3. Run and test the stack
4. Delete a specific model in the stack
The flow of stacks
Lets create a simple 2 AI model flow.
Input 1 is the start of your series of prompts, the AI model will create an output that then becomes part of the input of AI model 2.
In the example below, we are using the same AI model, first one to summarize the story. and then the same AI creates a polar opposite story from that information.
Imagine constructing a stack with diverse AI models, each specialized in understanding different sets of data. By leveraging multiple expert AI models, it becomes possible to process a series of data and generate solutions for complex situations. This approach enables the creation of robust solutions by harnessing the unique strengths of each model.
Was this article helpful?
That’s Great!
Thank you for your feedback
Sorry! We couldn't be helpful
Thank you for your feedback
Feedback sent
We appreciate your effort and will try to fix the article