4 ms·

Browser Agent Benchmark: Comparing LLM models for web automation

• pixel_popping 8mo ago
  It's lacking the best model (Opus 4.5) on the benchmark tho.
  • djohnston 8mo ago
    Yeah but then their own product might not score the highest.
    • pixel_popping 8mo ago
      Exactly why I'm pointing it out, which feels a bit corrupt, but understandable.
      • djohnston 8mo ago
        tbh i was a bit cranky yesterday - even if they are #2 on a legit benchmark that would be impressive
• wiradikusuma 8mo ago
  Since we're in this topic, can anyone suggest good AI-based tool for exploratory (fuzzy?) web testing?
• MagMueller 8mo ago
  [dead]