Introducing Extract API.
Fetch and parse any web page or PDF. 80% higher PDF accuracy on ParseBench-2.
→ Structured Tables, charts, forms, and scans
→ Try on web pages, no API key required to start
→ ZDR available for sensitive docs
Try it at https://t.co/iEABhfFHbJ
Building travel, ecommerce or finance agents?
With @youdotcom Knowledge, you can pull latest exchange rates, like USD to EUR from Xignite. Same $5 CPM.
Try it: https://t.co/uOhbhmiM6L
https://t.co/UscuspVhmd Web Search is now available in Qwen Code.
Install it and get access to a suite of tools right out of the box through a Keyless MCP integration.
100 searches per day, free to get started.
This past weekend, XDC Tech NYC proudly hosted Ground Truth: The Live Agents Hackathon, powered by @youdotcom!
Hackathons are where emerging technologies move beyond conversation into real-world innovation, bringing together the builders shaping the future of agentic AI.
Huge thanks to @somgupnyc, @BuilderSeries, @getopencrawl, @Shalin__Patel, @manishtyagi__, @Liv_Tarm, @celiasnetwork, @241_dom, @bondaicommunity, localhost:nyc, House of 434, Treg & everyone involved!
Congratulations to ElseWhere’s winning team: Khyati Thapliyal, Sachi Singh, Arnab Das & Naman Tyagi!
Better answers start with better context.
With @youdotcom Web Search built into Joshu by default, every response gets access to fresh and sourced information, without compromising on privacy.
Better context. Smarter decisions.
Joshu x @youdotcom: The upcoming release of Joshu will now feature the excellent https://t.co/6X7SEPdrxS web by default.
Fresh facts with sources, US-hosted, zero data retention.
Announcement in first comment!
$NVDA reported earnings last night. Does your agent know how the stock is doing right now?
Straight from S&P Global and Xignite, via Knowledge. Same $5 CPM.
Try it: https://t.co/7ULdjIU5US
Agents now use the web as much as people. But search was built for us, not them.
@saahil9jain, CTO of @youdotcom, on rebuilding search and inference for agents, and training open models to use it. AI Dev NYC main stage, Tuesday, Dec 1.
📅November 30–December 1, 2026 · Chelsea Piers, New York
🎟Get your tickets: https://t.co/kujsqE0oKo
#AIDev #DeepLearningAI #AIAgents
Introducing Extract API.
Fetch and parse any web page or PDF. 80% higher PDF accuracy on ParseBench-2.
→ Structured Tables, charts, forms, and scans
→ Try on web pages, no API key required to start
→ ZDR available for sensitive docs
Try it at https://t.co/iEABhfFHbJ
Qwen 4-Max is live on Inference Router, from Day 0.
The first model in @Alibaba_Qwen's Qwen4 series, built for coding, office work, and long-horizon agent runs. 1M-token context, native text, image, and video input.
Try it now: https://t.co/GW36Jlmf6N
Excited to bring @youdotcom's research capabilities to Circle Agent Marketplace!
One more way for developers to build smarter, more capable agents with reliable web intelligence built right into their workflows.
Give your agent research tools without building a custom integration from scratch.
Explore @trydrip, @MrScraper_, and @youdotcom on Circle Agent Marketplace for web search, page retrieval, publication content, and financial news.
Start with one research question. https://t.co/uUsPiWJOLq
Building macro or market research agents?
With @youdotcom Knowledge, you can pull the latest economic data, like US unemployment rate from Fred. Same $5 CPM
Try it: https://t.co/GQqoxlv2iR
What if you could build and ship an app without spending the day typing code?
We’re teaming up with @emergentlabs and @youdotcom for Speech to Ship, a one-day hackathon where you build using your voice.
November 20 in SF. No coding experience needed.
Link in comment
Updating RL policies every few hours creates an immediate bottleneck: getting new checkpoints onto inference workers without dropping serving availability.
Next Thursday, Nov 19th, engineers from @youdotcom, @NVIDIA and CoreWeave break down how to solve this in production today.
We’re walking through a live deployment of NVIDIA Nemotron 3.5 Lightning, using RL Rollouts and NVIDIA Dynamo on CoreWeave infrastructure to handle checkpoint hot-loading and on-policy weight transitions.
The direct outcomes on NVIDIA HGX B200:
✅ 22.94% accuracy gain
✅ 30.24% fewer tool calls
Get the exact blueprint, compute costs, and before-and-after traces. Register here: https://t.co/bQcQcevvH2