Cut search latency with native parallel AI search for Vertex AI

AgentTools now includes ToolParallelAiSearch, a native integration with Vertex AI's Parallel AI Search that allows agents to issue multiple search queries concurrently and aggregate results. This brings Vertex AI search into the same parallel retrieval pattern as other search tools, reducing latency for knowledge-intensive tasks that benefit from broad, simultaneous retrieval.

‍

Details:

ToolParallelAiSearch integrates directly with Vertex AI's native parallel search API
Enables concurrent multi-query search within a single tool call, reducing round-trip latency
Consistent with existing parallel search patterns in the toolkit; no special agent configuration required
Suitable for RAG workflows, research agents, and any use case requiring broad, fast retrieval from Vertex AI

‍
View the cookbook.