OpenChainBench
All products
Anthropic logo

Anthropic Benchmark

Anthropic is measured across 2 live OpenChainBench benchmarks: AI trading agents (ranks #4 of 8, -13.9% p50), AI trading agents (ranks #7 of 8, -26.9% p50). Data as of 2026-10-07 UTC.

1 benchmark

AI lab behind the Claude model family. Appears on OpenChainBench only through Recall Labs' trading arena, where Opus 4.5 and Sonnet 4.5 each drive one agent.

Last measured

Live benchmark results

  1. #4of 8

    Trading

    AI trading agents: what eight models earn net of the asset

    Alpha net of asset exposure

    -13.9%

    p50 · 24h

  2. #7of 8

    Trading

    AI trading agents: what eight models earn net of the asset

    Alpha net of asset exposure

    -26.9%

    p50 · 24h

Compare Anthropic to alternatives

No comparable providers yet.

Raw figures /api/citable