Browser Use
@browser_use
The AI browser agent
That's insane! ETH Zurich has just launched an AI Agent lab. They give builders the opportunity to create cool things while earning their degree. More universities should follow the great example set by ETH Zurich. We directly started to collaborate with them. Now you can do…
🚀Excited to share that @kev_osull , @PascalBtd , and I are launching the Agentic Systems Lab @ETH . We’re now assembling a Genesis Batch of 10–20 Master’s thesis students to build and evaluate agentic AI systems with real-world impact. If you want to spend your thesis shaping…
Even @grok approves @browser_use !!
Interesting integration! Browser Use could enhance the scraping step in n8n by handling dynamic web elements better than basic HTTP requests. For the leads finder, add a node: Use Browser Use to navigate Google Maps, extract data via selectors, then pipe to Hunter for enrichment.…
Who wants an MCP server for this? ⚡️ Like this post if you want it and we will ship it tomorrow 👨💻
❌ Do NOT use OpenAI for real-time web search - they index the web for fast yet inaccurate replies. 🚀 We’re releasing Browser Use Search API today. It crawls sites and fetches real-time data by interacting with any website. Built because you asked for it. We have already…
🏠🤖 Just built an apartment hunting bot using browser-use + @n8n_io ! It automatically: ✅ Scrapes multiple rental sites ✅ Finds best deals under $5k/month ✅ Sends formatted results to Slack ✅ Saves hours of manual searching SF rental hunting just got easier!
Agents > web search Agents can just login to check your balance. And top up?
Perplexity and OpenAI's web search are boring. They always stop where the data gets interesting. They stop at login walls. What is web search if you can't interact with the browser? Browser-use is the only API that can access interesting data behind login walls. We even do…
Voice assistant for any browser task. Works in your current chrome profile via an extension. made with: - @browser_use - electron-shadcn - some hacky CDP stuff
This is awesome!!!
Voice assistant for any browser task. Works in your current chrome profile via an extension. made with: - @browser_use - electron-shadcn - some hacky CDP stuff
🚀 Built an automated competitor research system using @n8n_io + @browser_use! Submit a form → get real-time competitor updates in Slack. No manual work. Want a step-by-step tutorial for similar use cases? cc: @sauravpanda24
Agent logs now available in the Cloud UI.
Wait… what’s that? 👀 FINALLY. You can now see exactly what went wrong with your agent run in @browser_use Cloud with logs and all. Yeah yeah… took us long enough 😂 But hey, better late than never, right? Try it out & debug like a pro. So, what do you want next? You can…
After months of requests… It's finally here! Browser profiles just dropped in @browser_use Cloud. Before: > every run used the same shared profile > no way to isolate sessions or logins > chaos with multiple accounts Now: > create multiple fully isolated browser profiles >…
Multi Browser Profiles live on the Cloud!
After months of requests… It's finally here! Browser profiles just dropped in @browser_use Cloud. Before: > every run used the same shared profile > no way to isolate sessions or logins > chaos with multiple accounts Now: > create multiple fully isolated browser profiles >…
Infinite creativity. A browser is powerful. It can do millions of things for you. Most people don't utilize that because they don't know how. We change that. You are one prompt away. You will just tell your devices your goals, and they do it for you.
Ran @browser_use on @ParadigmShiftAI to pit Claude 4 Sonnet vs Gemini 2.5 Pro on 10x10 WebVoyager vision tasks. Claude: 99 % accuracy & 3× faster ⚡️ Gemini: 75 % accuracy 😬 @GoogleDeepMind why the lag? #AI #VisionAI
Want an API where you send a search query and @browser_use agent goes into websites super fast in parallel and brings back real time info, all under 10 seconds? Comment everything you would do with this 👇
So like, we’re really on the edge right now—there’s a new developer role in town: AI x Web3. Prerequisites include: Interests in the following - Ownership of intelligence - @thinkagents - Distributed and decentralized inference contribution @rendernetwork @bittensor_…
If you build AI agents 1 step consumes 10k input tokens with less than 250 output tokens. So if you optimize for - speed: reduce output tokens - cost: reduce input tokens - reliability: don't just put more into context (performance can drop), but build clever systems around…