Monthly recap from Tavily:
@rweiss57 spoke at Nebius Inflection on the shift to two internets, one for humans, one for agents, then joined the Nebius for Startups Podcast.
New case study: Fieldway cut research time 91.7%, turning a 5-hour manual workflow into a 25-minute agent run with verified citations.
Shipped: keyless access, x402 pay-per-call, file uploads for /research, and new integrations with Snowflake, Databricks, Microsoft Copilot Studio, OpenAI Codex, and Cartesia.
Showed up at Snowflake Summit, Databricks Data + AI Summit, and community events with @LangChain and @TryArcade in Dallas and Atlanta.
More below + our first events in Europe this summer.
https://t.co/ip0DBIrJ2X
This past week, Tavily + @LangChain showed up in Dallas together π€
Two days with two formats - built around the people shaping what comes next.
Tavily and LangChain have been woven together since the start and we're excited to bring the people building with both into the same room.
When AI agents start making the infrastructure decisions humans used to make, your go-to-market changes completely.
@rweiss57, Founder & CEO of @tavilyai, on how Tavily reached 1 million users without outbound sales and why the winners in AI now are the ones cutting latency and token costs, not shipping the best model.
Hear the podcast: https://t.co/DfBR9piM6o
Tavily MCP Server is on @databricks Marketplace, connecting your agents to the live web through a secure Unity Catalog proxy. Search, extraction, crawling, mapping - structured for machine reasoning.
We're at Databricks Data + AI Summit this week. Come find us at our booth #435 to see it live, or learn more here: https://t.co/YYbQp2T5hi
Tavily is available on x402, which means an agent can call https://t.co/R5bMWRsdZ7, receive a 402 Payment Requiredresponse, pay $0.01 in USDC, and get search results back.
You donβt need an API key, account, or human configuring access in advance. The agent discovers the tool, pays for it, and keeps going.
If you're building autonomous workflows where tool access needs to happen at runtime, it's worth a look.
https://t.co/l3s7FKzgPb
Agents already live in the terminal. When they need web data, they should be able to run a command, so that's exactly what the Tavily CLI lets them do.
Search, extract, crawl, and research all return structured output that agents can parse directly, with no HTML to clean up or additional scraping in between.
One line to install, and your agent has real-time web access built into whatever loop you're running.
https://t.co/dXx9R8q8K0
Tavily Deep Research is a single API endpoint that runs multi-step research end-to-end, returning a structured, source-cited report instead of a list of links to sift through.
Pass in a question, get back a full research report. You can attach your own files alongside the web search, define an output schema so the response fits directly into your pipeline, and choose between mini for fast focused queries or pro for broad multi-angle topics.
Domain filtering, citation format, and output length are all configurable per request.
https://t.co/O3SqDhW9ka
We built a Discord for developers working with Tavily. Share what you're building, get your questions answered by the team, and connect with others doing the same.
Join here π https://t.co/9Qe8AzA9Xa
We're excited to announce our partnership with @coinbase to bring Tavily to x402, the open protocol for internet-native agentic payments.
With x402, agents can discover and use Tavily web search at runtime without an API key. Agents use a @Base wallet to pay per-request and get instant results.
The next billion agents will discover, pay for, and use online services fully autonomously. Weβre live on https://t.co/Cne0ZiUcpH, and weβre just getting started. More in the comments.
Quick demo on how my AI assistant Ava built a competitor research system using @tavilyai for web search + @vercel to deploy the findings
I have been building these "automations" for years now, but what's really interesting in this new "personal AI" world is that Ava starts to feel like a true colleague who understands our team, our market, and what matters to us
1/ She has more info about our competitors than any of us - so if we ask her anything in Slack she's going to be 10x more useful than any other ChatGPT or google search
2/ She's a great coworker: She has context on our work and is intentional about how she presents this data in a way that will matter more to us. It really feels like we have a 5th person in the marketing team now
personalized all the way π
Is that too long to wait? We're here with @nebiusai, @n8n_io, and @Base44 for our builders and brews event tomorrow, and there's still room! You can sign up now and we'll see you there!
https://t.co/JLqOu48fB0
Berlin was geht ab, Tavily ist jetzt in town! We're here with @GradiumAI showing off our new voice integration and hosting a hackathon alongside @nebiusai and @cursor_ai. You won't want to miss this one.
https://t.co/G6l8wxgZVT