Running Gen AI and LLMs on Cisco UCS X-Series v1

Learn how you & your customers can deploy the LLM (Large Language Model)/GenAI (Generative AI) software stack.

Overview

In this lab, you get exclusive access to a dedicated Cisco UCS X-Series hardware pod with the X210c M7 compute node and NVIDIA GPU.

• Deploy the Generative AI software stack.

• Explore and test AI models and inference servers.

• Perform retrieval augmented generation (RAG) & Stable Diffusion
  image generation.

Important Notes

• The tools and automation used in this demonstration are open-sourced, allowing your customers to try the same on their newly purchased UCS-X GPU-enabled system. Please refer to the demo guide appendix for details.
• This lab has limited capacity Please book ahead and keep your session duration to a minimum.

Get Started

1. Review the lab profile.
2. Click the Resources tab to view the documentation.
3. Scroll through the introductory topics to learn more about the solution, and remember to go back to the demo guide before you start your session.
4. At top right in the lab profile, click the Schedule button, and select the date and time for your lab session.
5. You can find your list of scheduled sessions in My Hub from the dCloud Home page.
6. When it's time, launch your session and consider Value Winners for your customers, as you explore the scenarios.