title: "Nimble launches Web Search Agents with 21-point accuracy gain" slug: "nimble-launches-web-search-agents-with-21-point-accuracy-gain" published: "2026-07-30" beat: "Launches" tags: ["Launches"] creator: "Agentry Newsroom" editor: "Susanne Sperling, Editor — Human in the Loop" tools: ["Claude (Anthropic)", "Perplexity Sonar"] creativeWorkStatus: "verified" dateReviewed: "2026-07-30" aiActArticle50: "compliant" humanView: "https://agentry.news/launches/nimble-launches-web-search-agents-with-21-point-accuracy-gain" agentView: "https://agentry.news/agent/nimble-launches-web-search-agents-with-21-point-accuracy-gain"
Nimble announced Web Search Agents on July 29, 2026, a retrieval system designed to improve web research precision for AI agents while cutting token consumption by 51% versus competing alternatives, a
Drafted by an AI agent. Verified by Susanne Sperling, Editor — Human in the Loop. AI policy.
Nimble, the web search platform for AI agents, launched Web Search Agents on July 29, 2026, claiming significant improvements in retrieval accuracy and efficiency for AI-driven research workflows VentureBeat.
The new product is positioned as domain-specialized retrieval infrastructure. According to Nimble's announcement, the agents "autonomously learn the user's domain and adapt retrieval strategies to each task, so each agent gets the exact web context it needs" TechTarget.
Nimble's own benchmarks, which the company notes were "conducted by Nimble but not independently verified," claim a 21-point increase in answer quality and 51% fewer tokens spent per query versus leading AI search alternatives TechTarget. The company frames Web Search Agents as a departure from the status quo: "Until today, AI agents all had the same web search," according to Nimble's announcement materials.
These performance metrics address a core pain point in the agent economy—the cost and latency of web-based retrieval for AI systems that need external context to complete tasks. Token efficiency matters directly to operational costs for enterprises running large fleets of agents.
Nimble is offering access through a pay-as-you-go Agent API priced at $0.025 per request and managed plans starting at $2,500 per month, according to coverage by VentureBeat. The tiered pricing model targets both small-scale developers and enterprise deployments.
The launch follows Nimble's $47 million Series B funding round in February 2026, led by Norwest Nimble, positioning the company to compete in the rapidly consolidating agent infrastructure market where retrieval quality and cost directly influence agent performance at scale.
Web search quality has become a critical differentiator in the agent economy. As agents take on more autonomous decision-making and research tasks, the relevance and cost-efficiency of retrieval systems determine whether those agents remain practical for enterprise deployment. Nimble's announcement suggests the market is moving beyond generic search APIs toward specialized retrieval optimized for specific domains and agent use cases.