Italiano
Gregor Zunic
Segnala

Gregor Zunic✓

@gregpr07

founder @browser_use

611 Seguiti31.8K Follower3.9K Post
  1. Gregor Zunic✓@gregpr07
    0. browser use - 5 employees
    Jared Cohen @jaredccohen
    we need a (mindshare/headcount) index for startups. some leaders are probably: 1. cal - 30 employees 2. monaco - 90 employees 2. granola - 120 employees 3. parallel - 125 employees 5. exa - 180 employees ramp also scores well despite their denominator being 2000
    007551
  2. Jared Cohen✓@jaredccohen
    we need a (mindshare/headcount) index for startups. some leaders are probably: 1. cal - 30 employees 2. monaco - 90 employees 2. granola - 120 employees 3. parallel - 125 employees 5. exa - 180 employees ramp also scores well despite their denominator being 2000
    40468.9K
  3. Gregor Zunic✓@gregpr07
    Browser Use stealth is the best in the gane
    Aitor Mato @reformedot
    While others get flagged as bots, @browser_use solves all major anti-bots and CAPTCHAs. SOTA on every public stealth benchmark. Fully headless, with in-house captcha solvers. Now I'm only competing against myself. Blocked somewhere? DMs are open. Blog post soon. 🤓
    118484
  4. Gregor Zunic✓@gregpr07
    Scarica
    Lukas Gruber @lukasgruber__
    Instinct is currently raising @ $30b. $100M → $500M → $2.5B → $10B → ~$30B in sub 1y. If you're still not inside the private market this should be a good signal to pivot aggressively. Wealth creation is going to be heavily skewed toward private investors over the coming decade.
    001730
  5. Lukas Gruber✓@lukasgruber__
    Instinct is currently raising @ $30b. $100M → $500M → $2.5B → $10B → ~$30B in sub 1y. If you're still not inside the private market this should be a good signal to pivot aggressively. Wealth creation is going to be heavily skewed toward private investors over the coming decade.
    149311.4K661K
  6. Gregor Zunic✓@gregpr07
    We spent months with CMU building the benchmark we actually wanted > 100 long tasks from what people really automate > best agent fully completes 47/100 > a stricter judge cuts average scores in half evals are the bottleneck for agents now
    Browser Use @browser_use
    Introducing: Odysseys2 - hardest long horizon benchmark for web agents based on real user tasks 🏗️ We collaborated with @JangLawrenceK @rsalakhu from Carnegie Mellon University > SOTA agent fully completes only 47/100 > 5 agents complete none > 8-12 real rubrics per task, 60 min on the live web Web agents are becoming better than humans at knowledge work.
    207635
  7. Browser Use✓@browser_use
    Introducing: Odysseys2 - hardest long horizon benchmark for web agents based on real user tasks 🏗️ We collaborated with @JangLawrenceK @rsalakhu from Carnegie Mellon University > SOTA agent fully completes only 47/100 > 5 agents complete none > 8-12 real rubrics per task, 60 min on the live web Web agents are becoming better than humans at knowledge work.
    Scarica
    43282.4K
  8. Gregor Zunic✓@gregpr07
    Gregor Zunic
    RT @reformedot: While others get flagged as bots, @browser_use solves all major anti-bots and CAPTCHAs. SOTA on every public stealth benchmark. Fully headless, with in-house captcha solvers. Now I'm only competing against myself. Blocked somewhere? DMs are open. Blog post soon. 🤓
    010132
  9. Gregor Zunic✓@gregpr07
    You can just one shot the stuff you always wanted these days
    Gregor Zunic @gregpr07
    I wanted both Studio Displays to be a stereo speaker, so my agent one shotted this in Rust and launched this post. > native controls work > native menu bar > starts at login open source. Rust + Swift.
    1051.1K
  10. Gregor Zunic✓@gregpr07
    I wanted both Studio Displays to be a stereo speaker, so my agent one shotted this in Rust and launched this post. > native controls work > native menu bar > starts at login open source. Rust + Swift.
    Scarica
    1082.1K
  11. Gregor Zunic✓@gregpr07
    Gregor Zunic
    RT @browser_use: GPT-6.1 Terra is insanely good at browser use 🔥 > 78.5 on the Browser Use Benchmark > basically Sol performance (79.7) at 35% less cost. Pareto frontier is once again dominated by OpenAI.
    0220134
  12. Gregor Zunic✓@gregpr07
    Gregor Zunic
    RT @browser_use: Kimi K3.1 Pro is a beast at browser use 🤯 76.8 on the Browser Use Benchmark. Only GPT-6.1 Sol scores higher. It beats Opus 5.5, Grok 4.7 and Sonnet 5.5, and it's cheaper than Opus at ~$2/task. Best Chinese model we've ever tested, by a lot. Next best is DeepSeek V4.1 Flash at 70.6. Moonshot cooked.
    080126
  13. Gregor Zunic✓@gregpr07
    Kimi K3.1 Pro is the best open source model for long horizon knowledge work.
    Browser Use @browser_use
    Kimi K3.1 Pro is a beast at browser use 🤯 76.8 on the Browser Use Benchmark. Only GPT-6.1 Sol scores higher. It beats Opus 5.5, Grok 4.7 and Sonnet 5.5, and it's cheaper than Opus at ~$2/task. Best Chinese model we've ever tested, by a lot. Next best is DeepSeek V4.1 Flash at 70.6. Moonshot cooked.
    100262.6K
  14. Gregor Zunic✓@gregpr07
    Gregor Zunic
    RT @Alezander9: Introducing: Mainplane🛫 A new open-source AI agent I built from scratch in Go Local agents die with your laptop. Cloud agents can't reach your work. Mainplane lives on your server and drives all your computers > Your old computers → always-on AI shed > Shared file server for all computers > Use a Claude/ChatGPT plan on regular usage limits > Browser + computer use on Mac, Windows and Linux One line to install 100% open source, MIT ↓
    0230197
  15. Gregor Zunic✓@gregpr07
    It’s insane that grok bot gives access to X api for free Scraping all my 31k followes would cost $300 via api, my grok bot did it for free
    531295355K
  16. Gregor Zunic✓@gregpr07
    Small teams move 100x faster
    82453.6K
  17. Gregor Zunic✓@gregpr07
    As a startup founder, you can’t afford shame
    909564922.2K
  18. Gregor Zunic✓@gregpr07
    Duck-tape engineering is the way to build startups.
    60453.2K
  19. Gregor Zunic✓@gregpr07
    I needed this 5 years ago. I don’t even have an IDE now :(
    colinhacks @colinhacks
    type inference from .sql files has been achieved internally ORMs are toast
    729823.7K
  20. colinhacks✓@colinhacks
    type inference from .sql files has been achieved internally ORMs are toast
    Scarica
    2073565.4K388K
  21. Gregor Zunic✓@gregpr07
    Introducing: Open Duet
    Scarica
    Magnus Müller @mamagnus00
    I just launched Open Duet. Your agents learn what matters to you, find useful work, and prepare it for your approval. Run it with Claude Code or Codex.
    599910.3K
  22. Magnus Müller✓@mamagnus00
    I just launched Open Duet. Your agents learn what matters to you, find useful work, and prepare it for your approval. Run it with Claude Code or Codex.
    Scarica
    1244812.3K
  23. Gregor Zunic✓@gregpr07
    Are you making Muse/Instinct/Boba competitor? Our cloud browsers can bypass all antibots and browse the web like a human.
    Garry Tan @garrytan
    Aside Browser is a real solution to making your agents actually great at using the web. @AsideAI
    32923218.5K
  24. Garry Tan✓@garrytan
    Aside Browser is a real solution to making your agents actually great at using the web. @AsideAI
    Scarica
    501422448.7K
  25. Gregor Zunic✓@gregpr07
    Post @gregpr07
    004986
  26. Gregor Zunic✓@gregpr07
    Gregor Zunic
    RT @humansand: Introducing Boba - the social intelligence that understands your goals and your community. Text Boba today or message it online at
    0390203
  27. Mostra altri post