"An enterprise multi-agent orchestration stack from OpenAI which will best fit internal Chat based workflows."

OpenAI introduces AgentKit, a new suite of tools for building, deploying, and improving agentic workflows end-to-end. It combines a visual Agent Builder for designing and versioning multi‑agent flows, a Connector Registry to centrally manage and govern connectors across OpenAI products (including third‑party MCPs), and ChatKit for quickly embedding customizable chat UIs with streaming and thread support. The announcement also expands Evals with datasets, trace grading, automated prompt optimization, and third‑party model evaluation to help teams measure and iterate on agent quality. Finally, it highlights reinforcement fine‑tuning for reasoning models and outlines current availability: ChatKit and the new Evals features are generally available, Agent Builder is in beta, and Connector Registry is rolling out to select enterprise customers under standard API pricing.