You can host almost any LLM on watsonx! In this video I will walk you through, how you can do this for deepseek-r1 (distilled-llama-8b). Follow the official documentation if you want to know all the details: timestamps: 00:00 - Intro 01:37 - The official Documentation 02:24 - Prerequisites 03:13 - Downloading deepseek-r1-distill-llama-8b from huggingface 06:35 - upload the model IBM Cloud using Aspera high-speed transfer 08:50 - connect the storage to watsonx 10:49 - register the custom foundation model in watsonx 12:35 - deploying the model 13:22 - using deepseek in the prompt-lab 14:27 - outro