Absolutely this is a strong milestone for Sai and Simular. Here’s a professional, engaging quote-tweet text:
A major milestone for AI agents 🚀
Sai is now leading the OSWorld 2.0 leaderboard with a 79.38% partial success rate, outperforming some of the biggest labs in real-world computer-use environments.
This is a great example of how reliable autonomous computer execution is becoming a reality. 🔥
Definitely worth keeping an eye on @sai_borg and @SimularAI.
It’s official 🏆 Sai is on OSWorld 2.0!
@sai_borg, our model-agnostic computer-use agent, is #1 in partial success rate (79.38%) across real-world desktop environments.
We didn’t just beat the bigger labs; we defined a new way to build autonomous computer execution: reliable and cost-efficient.
Cost per task of Sai = $14.34
Anthropic's Claude Computer Use with Opus 5 = $24.11