Step into the world of cutting-edge AI innovation in this session, where you’ll experience the powerful synergy between InstructLab and Red Hat OpenShift AI (RHOAI). Together, we’ll explore how to fine-tune open-source models using InstructLab and seamlessly bring them to life in an OpenShift (k8s) environment on RHOAI.
We will help you discover the magic of ilab and distributed workloads with RHOAI both at the same time, by enabling faster and more efficient training for even the most demanding AI models. You’ll dive deep into an end-to-end training journey—starting with generating synthetic data (SDG) to enhance your model’s capabilities on ilab, fine-tuning IBM’s Opensource Granite model in a phased approach on OpenShift AI, and culminating in a rigorous evaluation process using Mistral-Eval as the judge with MTBench.
This session isn’t just about tools and workflows; it’s about empowering you to tackle real-world AI challenges with confidence and creativity. Start small with RHELAI (ilab) and then expand your MLOps journey on RHOAI with seamless integration. Join us to unlock the full potential of OpenShift AI and take your AI training and experimentation to the next level.