Apple Watch Running GPT2 Locally

GPT2 (distil) running locally on the Apple Watch Ultra 2 with Swift CoreML Transformers.

Todo:

Figure out why the app crashes after certain # of tokens are generated - running out of memory? The GPT2 Model is nearly 500 mb. I believe the Apple Watch Ultra has 1 GB of RAM

Resources:

Swift CoreML Transformers Repository - https://github.com/huggingface/swift-coreml-transformers