A demo session from Fireworks AI showing how to deploy and evaluate open source LLMs on Microsoft Foundry. Covers the end-to-end workflow: discovering models, comparing them side-by-side in a playground, running evaluations with custom datasets, creating agents, and moving from testing to production via dedicated or multi-tenant deployments. Also touches on bringing custom fine-tuned model weights to Foundry for optimized inference using the Fireworks serving stack.
•14m watch time
1 Impression