1 comment

[ 2.6 ms ] story [ 12.2 ms ] thread
A web agent that autonomously navigates websites on its own cloud browser to complete tasks for you.

Navigator achieves pareto-domination over Gemini 2.5, Claude 4.5, and OpenAI Operator

  • 10%-20% accuracy gains across benchmarks
  • 2-3x faster
  • Uniformly preferred in head-to-head human-evals
(Gemini 3 computer-use hasn't been released yet, so no comparison possible)

Navigator is trained with mid-training, SFT, and RL — RL not only on simulated web envs, but also direct interactions with live websites!