AIHacker Newsabout 3 hours ago
Surge AI launches DAYJOB Tax benchmark
7 min read
Surge AI introduces DAYJOB: Tax, a benchmark evaluating AI agents on 50 professional tax assignments. Claude Sonnet 5.5 leads with a 10.8% pass rate, while Opus 5.5 achieves 9.2%. The suite covers eight practice areas and contains 1,789 workspace files, including 1,487 PDFs and 117 Excel workbooks, testing agents on complex judgment and evidence resolution.
Level
Hype check
Tap to vote and see what everyone thinks.
