We’re releasing a more advanced speech-to-speech model and new API capabilities including MCP server support, image input, and SIP phone calling support.
