Skip to content
Slow Variables

Ask the data

Answers come from the same store as the site; every number is checked against the record it cites.

Indicators / series

epoch_bench.claude_sonnet_4_20250514_8k.arc_agi_2.pt

Source: Epoch AI benchmarks and Capabilities Index (Epoch AI) · CC BY 4.0 · CSV

Epoch AI, Benchmarking Hub and Epoch Capabilities Index (CC BY 4.0)

idvaluegradeas of datepublished datetieraudited vs reportedextraction methodreview statusretrieved athttp statuscontent hashsupersedes idurlflags
992d6c5ef753dcdf2.1%grade B, evidence tier 62025-05-222025-05-226reportedapiapproved2026-09-10T06:33:11.514623Z200cde2f74ecf00source page
Raw snippets and dispute text
  • 992d6c5ef753dcdf {"Cost per task": "0.2654", "Model version": "claude-sonnet-4-20250514_8K", "Name": "Claude Sonnet 4 (Thinking 8K)", "Organization": "Anthropic", "Release date": "2025-05-22", "Score": "0.0212"}