In this lab, you will have access to a deployment of an AMD blueprint for a Financial Stock Intelligence Agent. It is deployed on AMD Instinct MI250X instance in the ATC. Built using LangChain and powered by AMD Inference Microservices (AIM), this solution combines real-time market data, technical indicators, and large language model analysis to deliver AI-generated stock insights through an interactive web interface. Financial analysts and developers can explore how AMD's open AI stack accelerates inference for financial workflows.

In this lab, customers explore the LLM Router deployed on AMD Instinct MI250 accelerators in the ATC. Built from the AMD Enterprise AI Suite solution blueprint catalog, the LLM Router puts a single OpenAI-compatible endpoint in front of a portfolio of models and decides, prompt by prompt, which model should answer. Lightweight requests are served by a small, fast model; hard reasoning and code generation are escalated to a large one. The result is a practical answer to the accuracy-versus-cost tradeoff that every organization running self-hosted inference eventually confronts — delivered on a fully open source AMD software stack.