Benchmarks / Web research

Web research

Find hard-to-locate facts and answers on the web.

• BrowseCompby OpenAITask level · Accuracy

  92.2%GPT-5.6 Sol Ultra · Jul 2026

  Saturated
• Finance Agent Benchmarkby Vals AITask level · Accuracy

  64.4%Claude Opus 4.7 · 4 Jun 2026

  Retired
• GAIAby Meta and Hugging FaceTask level · Test-set accuracy

  93.7%Ops-Agentic-Search-2.0 · 22 Sep 2026

  Saturated
• Legal Research Benchby Vals AITask level · All-pass accuracy

  55.3%Muse Spark 1.3 Max · 22 Sep 2026

  Emerging
• MCP Atlasby Scale AITask level · Pass rate

  88.1%Muse Spark 1.1 · 9 Jul 2026

  Strong
• Tax Agent Benchby Vals AITask level · Accuracy

  49.2%Claude Fable 5.1 · 8 Oct 2026

  Emerging