Introducing Agent Ultra - a step change in deep research Agent Ultra orchestrates swarms of agents to perform exhaustive research, build comprehensive lists, and answer questions requiring thousands of sources. In both evals and vibes, it's state of the art

Jev with Exa is INSANE. > Jev without websearch confidently gives wrong outputs > Jev with websearch is literally much more accurate Try Jev (with Exa websearch) for free 👇

We built a time machine for the web. Introducing Exa Snapshot: an index of 400 billion historical snapshots of webpages that lets you search as if it's the past. Snapshot is already being used for backtesting prediction models, RL at labs, exploring the pre-AI web, and more.

We’re releasing a research preview of “dynamic highlights”: Given a query and set of pages, a custom model jointly scans the whole set to filter out redundant and irrelevant info, reducing token usage 95% on average. Available today:

Exa is partnering with @firefox to power search for an AI browser for the open internet. @WilliamBryk sat down with Ajit Varma, Head of Firefox, to talk about the future of the browser and why user choice matters more than ever.

Exa is now live on the Machine Payments Protocol (MPP), the open standard from @tempo and @stripe. Agents shouldn't need accounts or API keys to search the web! With MPP, agents can pay per request with USDC.e on Tempo.

Exa is now one of the largest indexes in the world. We serve 80B pages, track 1.4T urls, and are on track to be Google-scale in early 2027. It's difficult to know the size of indexes, but we estimate Google to be ~1T, Bing to be ~500B, and Yandex ~200B. Brave mentioned in April that they're at 40B. Most of the web is trash that can hurt AI outputs, so you actually have to crawl way more than you serve and train models to filter out the trash. In terms of peak QPS, we estimate Bing at roughly 30k/s and Google at ~500k/s. In our case, because agent traffic often requires large fan-out (deep searches can use dozens of sub-searches) and can be spiky (like when AI labs RL with us), we're already starting to provision for Bing-scale traffic. Within 2 years, agents will search at many millions per second. They'll also want comprehensiveness over all data, far bigger than any index today. So the retrieval infra required will be larger than Google-scale in both dimensions.

We were happy to see a 3rd party evaluator compare Exa vs native search. On their benchmarks, adding Exa to Fable, Sol, Grok, or Gemini increases accuracy across domains, with the biggest jump in finance. Just the beginning!

Agent and Exa Connect are now live in our MCP! Find any data both on the web and from providers like Similarweb and Zoominfo directly within Claude and Codex

Introducing Exa Connect: connecting agents to data beyond the public web. Available today with ZoomInfo, Crunchbase, Similarweb, and many other leading data providers.

How does Exa compare to Google for training LLMs to search? In this blog post, we find that LLMs using Exa during reinforcement learning reach higher performance with 70% less training compute.

The Exa MCP is now officially available in ChatGPT! Exa gives ChatGPT access to unique data sources like people, companies, papers, and more. We find it particularly useful inside of Deep Research. To install, search “Exa” in “Apps”

We're excited to help CodeRabbit reduce the number of bugs in the world! By using Exa search, the CodeRabbit review agent can ground code reviews with technical information better, 5 times faster.

A search engine has many moving parts, and they constantly change with scale of index and traffic. In this blog post, we dive into the code-level abstractions we've built to develop search infrastructure that is both performant and a joy to work with.