Glean Introduces Specialized Agentic Search Model Waldo for Enterprise AI Efficiency

A LinkedIn post from Glean highlights the introduction of Waldo, described as the company’s first agentic search model designed to orchestrate enterprise search workflows alongside large frontier models. According to the post, Waldo plans multi-step queries, selects tools, and determines when sufficient evidence has been gathered before handing off to a more general-purpose reasoning model.

The post suggests that Waldo targets retrieval-heavy use cases such as analysis, content creation, and action-taking, where latency and cost are significant constraints for enterprise AI deployments. Glean reports that on a per-LLM-call basis, Waldo delivers roughly 10x faster performance, with about 250 ms P50 latency versus around 3 seconds for its default reasoning model.

In Glean’s internal harness, the LinkedIn post cites roughly 50% lower latency and about 25% lower token cost, implying a potential path to more cost-efficient and responsive AI-powered search for enterprise customers. The model is described as specialized for search planning, building context first and deciding whether a query requires a rapid response or deeper reasoning, which could make AI usage more predictable at scale.

The post also notes that Waldo is built on NVIDIA’s Nemotron 3 Nano and has been post-trained for search planning, with NVIDIA and Thinking Machines Lab referenced as partners in its development. This collaboration may indicate closer alignment with leading AI hardware and model providers, potentially strengthening Glean’s competitive positioning in enterprise AI search and improving its ability to attract large customers seeking lower-latency, lower-cost solutions.

For investors, the emphasis on specialized models for high-demand, well-defined jobs suggests a strategic focus on operational efficiency and scalability rather than solely on general-purpose frontier models. If Waldo’s reported performance and cost metrics translate into customer adoption and higher usage, Glean could see improved unit economics and enhanced differentiation in a crowded enterprise AI market.