Deploying a Text-Generation Inference server on a Google Cloud TPU instance

Stay tuned!