Introducing Linkup Flash (Alpha): the most accurate web search for agents, in under 200 milliseconds

Flash is available through one parameter:

const response = await client.search({
  query: "Did AI solve Navier-Stokes?",
  depth: "flash",
  outputType: "searchResults",
});

Pushing the Pareto frontier at sub-200ms latency

Flash creates a new point on the accuracy–latency frontier: among the APIs we measured, none is both faster and more accurate.

200 milliseconds is the natural gap between turns in human conversation (Stivers et al., 2009). A search call that completes within it can run inside a live exchange without creating a pause the user notices. We treat it as the latency budget retrieval must meet to disappear into a conversation.

We benchmarked Flash against the fastest search mode of each major web search API on SimpleQA, under identical conditions: the same queries, the same number of results, latency measured client-side. We measured each provider from four cloud regions (us-east4, us-central1, us-west2, europe-west1) and report its best region.

Latencies measured this way can exceed providers' published figures, which are typically produced under conditions each provider chose.

Flash returns results in 190 milliseconds at the median and 295 milliseconds at the 90th percentile — the most accurate web search API on SimpleQA under 200 milliseconds.

And the point will move as we keep improving accuracy at this latency.

From search orchestration to intelligent retrieval

The first generation of web-augmented AI systems delegated most of the work to the search layer: query decomposition, retrieval planning, source selection, page reading, answer synthesis.

As AI systems take on mission-critical tasks, teams are bringing that work inside the application to keep control over the output. Planning, tool selection, iteration, and evaluation now happen in the agent itself: it decides what to retrieve and how each piece of evidence shapes its next action.

As orchestration moves into the application, the interface with search becomes narrower and more demanding: relevant, clean, AI-ready evidence, delivered in milliseconds.

The same shift is underway at the model layer. As reasoning and tool use improve, system quality depends less on what a model memorized during training and more on reliable access to the right information at inference time. Current knowledge cannot live in model weights alone.

Flash is built for this interface: reliable web evidence, as fast as we can serve it.

Built for high-throughput, low-latency applications

At 200 milliseconds, Flash is built for latency-critical applications:

• Real-time voice and embodied systems. Voice agents, wearables, vehicles, and robots run complex workflows inside budgets of a few hundred milliseconds. At 200 ms, web search fits inside a single conversational turn.
• Deep research. These systems work in rounds — search, read, search again — so retrieval latency compounds. Cutting it buys more rounds in the same time: more leads explored, better sources read.
• Reinforcement learning. Thousands of parallel training episodes each make their own searches, so search latency sets the pace of training itself. Faster calls mean more training on the same hardware.
• Small language models. Small models trade stored knowledge for reasoning and tool use — a trade that only works if external knowledge arrives fast. Flash keeps small and local models viable for knowledge-intensive tasks.

Low latency starts at the index infrastructure level

Flash holds retrieval quality at this latency because it runs on Linkup's own index and search infrastructure: we control the full path from source selection and refresh to document processing, retrieval, and ranking. The speed comes not from any single component but from optimizing storage layout, retrieval stages, and ranking models together, and evaluating them under the conditions they face in production.

This efficiency makes the stack portable — across regions, inside a customer's cloud, or on-premises — so serving infrastructure can be placed close to the client, and the same ultra-low latency holds anywhere in the world.

Search systems designed around the application

Most web search APIs expose a fixed system: one index, one ranking pipeline, one deployment model. We think the knowledge layer of an AI system should be designed around the application it serves.

Operating our own index makes that practical:

• Client-specific indexes and retrieval pipelines. The source corpus, refresh policy, ranking models, retrieval stages, and query capacity can each be configured per application.
• Deployment that follows security and privacy requirements. Zero Data Retention, regional processing, private cloud, and on-premises installations.

Teams control both what their AI can retrieve and where that retrieval takes place.

Try Linkup Flash (Alpha)

Flash is available in alpha through the Search API:

depth: "flash"

Try Flash in the playground, read the Search API documentation, or talk to our team about dedicated capacity, a tailored index, or a private deployment.