Nimble Launches Expert Web Search Agents That Improve the Quality of Results and Reduce Token Spend

Independent benchmarks show a 21-point jump in answer accuracy and a 51% reduction in token costs compared to leading AI search tools

NEW YORK, July 29, 2026 (GLOBE NEWSWIRE) -- Nimble, the expert-level web search platform for AI agents, today announced Web Search Agents, which autonomously learn the user’s domain to execute complex web research tasks accurately and efficiently. Nimble’s web search, crawling, and research agents are specialized in self-learning for your use case, driving accurate, token-efficient usage where generalist web search falls short. Nimble delivers the exact web context agents need. Benchmarks against leading alternatives show a 21-point increase in answer quality and 51% fewer tokens spent per query.

Nimble’s AI lab research breakthrough unlocks a new category of web search infrastructure that enables AI agents to operate with the depth, precision, and context of a domain expert rather than a generic web search engine. Generic web search tools offer the same product for market research agents and lead enrichment agents, returning a wide, unstructured set of results and leaving the agent to sort out what is actually relevant. Using general web retrieval burns tokens by performing unnecessary tool calling and processing irrelevant content.

Nimble expert-level search also delivers higher accuracy for mission-critical AI, unlocking use cases where generic web search is either not accurate enough or too costly. The specialized harness self-learns the knowledge work and adapts retrieval strategies to each task. It combines proprietary indexes and real-time web retrieval for these agents to uncover the freshest, most complete, and most granular information across the web.

“For enterprises, the real bottleneck is accuracy and cost,” said Uri Knorovich, CEO and co-founder of Nimble. “Agents need the right live web context without wasting tokens or relying on generic search results. Nimble combines self-learning retrieval with deep search expertise to deliver higher accuracy at lower cost, unlocking workflows where generic web search falls short.

Expert-Level Search Infrastructure for Production AI Agents
For business-critical workflows, long-running agents, accuracy and completeness matter more than saving a few seconds of latency. Nimble’s agent harness allows agents to reason, gather, validate and synthesize information more effectively, producing higher-quality insights and reducing the manual research burden on teams.

The business impact of this adaptable approach is already measurable: Rox, an AI-native CRM company, reported a 20x reduction in token costs after adopting Nimble’s expert-level search infrastructure, while improving the quality and completeness of information delivered to its agents.

Nimble is now available via API, SDK and MCP, giving any AI builder direct access to live web intelligence, from low-latency web search to deep research and structured dataset generation from the web. Developers can connect Nimble as a tool inside their agent or build end-to-end applications powered by accurate, complete and traceable web context. Nimble is already running in production with leading Fortune 500 enterprises and AI-native companies, powering more than 90 million searches daily for mission-critical workflows where accuracy, completeness, and enterprise control matter.

“What impressed me most is how Nimble enabled us to tune a Claude Managed Agent to surface the competitor signals that actually matter to our team, rather than producing a generic market summary,” said Almog Lavi, Head of Product at Qodo.

Nimble Web search, crawling, and research agents are available via a free trial. Explore our latest benchmark analysis to learn how specialized web search agents outperform generic approaches when evaluated against industry-specific use cases instead of one-size-fits-all benchmarks.

Attending AI4? Book a personalized demo to see how Web Search Agents specialize in your industry and execute complex web research with expert-level precision.

About Nimble
Nimble delivers expert-level web search for AI agents, driving higher accuracy while cutting token costs. Instead of the generic, noisy responses returned by one-size-fits-all search tools, Nimble's Web Search Agents learn each of your use cases to retrieve exactly the information that's needed. This domain expertise compounds over time through a Proprietary Index and Memory, avoiding redundant searches and reducing reliance on LLMs to parse raw web pages.

Trusted by hundreds of enterprises and headquartered in New York, the company is backed by leading global investors, including Norwest, Databricks Ventures, Target Global, Square Peg, Hetz Ventures, Slow Ventures, R-Squared Ventures, J-Ventures, and InvestInData.

For more information, visit https://www.nimbleway.com/.

For Media Inquiries Contact:
BOCA Marketing Agency
nimble@bocamarketing.com

Legal Disclaimer:

EIN Presswire provides this news content "as is" without warranty of any kind. We do not accept any responsibility or liability for the accuracy, content, images, videos, licenses, completeness, legality, or reliability of the information contained in this article. If you have any complaints or copyright issues related to this article, kindly contact the author above.