Unlock the potential of AI reasoning on your local machine. In this webinar, we’ll show you how to leverage HP AI Studio to transfer DeepSeek’s advanced reasoning capabilities to deployable local models. Joseph Salzano, HP Software Architect and Engineer, will demonstrate these cutting-edge techniques that are transforming AI implementation.
Whether you’re an AI developer, data scientist, or an IT leader, you’ll gain actionable insights to elevate your organization’s AI capabilities.
What you can expect to learn:
- Understand DeepSeek’s Mixture of Experts (MoE) architecture and its advantages
- Techniques for running DeepSeek-R1 locally for inference
- How HP AI Studio streamlines model training using Dolphin-R1 datasets from Hugging Face
- How to enhance Llama 3.1 with chain-of-thought reasoning and deploy to Ollama for local inference
- See it in action locally on a Z8 Fury workstation with 128GB of RAM, three NVIDIA RTX
A6000 GPUs, one NVIDIA RTX A4000 GPU, and an Intel
Xeon processor with 44 cores
Speaker: Joseph Salzano, Software Architect and Engineer at HP