Skip to content
Slow Variables
Ask

Indicators / series

hal_reliability.claude_sonnet_4.accuracy.pt

Source: HAL AI Agent Reliability Tracker (reliability scores by agent) (Princeton University, HAL) · HAL (Princeton); cited with attribution · CSV

HAL: Holistic Agent Leaderboard, AI Agent Reliability Tracker, Princeton University

idvaluegradeas of datepublished datetieraudited vs reportedextraction methodreview statusretrieved athttp statuscontent hashsupersedes idurlflags
26427933ae35ae0866.5%grade A, evidence tier 12025-05-222026-09-221reportedscrapeapproved2026-09-22 10:28200a0da5842bdd0—source page
Raw snippets and dispute text
  • 26427933ae35ae08 {"label": "Claude Sonnet 4", "measure": "accuracy", "provider": "anthropic", "slug": "claude-sonnet-4", "x": "2025-05-22", "y": 0.6647630147630148}