Construct a chatbot with a Streamlit* front end using the power of LLMs. Connect your OpenAI*-compatible API and model endpoint to Hugging Face*. In this demonstration, the model inference endpoints are hosted on Intel® Gaudi® accelerators and deployed on Denvr Dataworks* cloud servers.

As technology advances, chatbots increasingly streamline processes, including generating code, writing blog posts, summarizing text, and conducting market research. Hugging Face Spaces offers a simple and flexible platform to build and share sample applications. Use Intel Gaudi accelerators with the Intel® AI for Enterprise Inference, an Infrastructure-as-a-Service (IaaS) product, hosted by Denvr Dataworks. Use your OpenAI-compatible API key and hosted model endpoint to get started.

The session explores numerous valuable AI tools and introduces environments that facilitate the essential steps in constructing a chatbot. Key technologies are showcased, including Python*, Streamlit, OpenAI API, and Hugging Face Spaces.

Key takeaways from this session include:

• Building an LLM-powered chatbot on Hugging Face
• Coding a front-end Streamlit application
• Using API secrets on Hugging Face
• Using a modern OpenAI-compatible API

The session is geared to novice developers.