DeepSeek R1 is the first fully open-source reasoning model that is on par with OpenAl's 01. Best-in-class AI can now be sovereign AI, running in private data centers and even air-gapped environments, at a fraction of the cost of leading LLM AI services.

Hopsworks makes on-premises deployment (and tuning) of DeepSeek easier than ever with an Enterprise grade platform for managing data and compute for AI. Hopsworks model deployments are built on KServe/vLLM, adding AAA security, observability, scale, GPU sharing, and ease of use.  

In this webinar we walk through how you can get DeepSeek models up and running in an on-premises or cloud-based Kubernetes cluster in minutes with Hopsworks.