It would be great for Amplify to support LLM in Predictions

**Describe the solution you'd like**
I want simple abstractions build-in in Amplify to work with LLM models like ChatGPT or AWS one when they release it.

**Describe alternatives you've considered**
There are open source projects, they are roller coasters.

**Additional context**
Even though LLM models have very simple APIs for inference, making useful apps is not very easy undertaking. If Amplify can offer such integration, it would be a huge boost to the ecosystem, I believe.