Surge AI releases DAYJOB tax agent benchmark
What happened
On October 28, Hacker News' front page reported that Surge AI launched DAYJOB: Tax, a benchmark for evaluating how well AI agents do professional tax work. It covers eight categories of tax work across 50 task environments, and a run passes only if it meets at least 95% of a task's criteria. The results listed: Claude Sonnet 5.5 scored 10.8% pass@1, Claude Fable 5.1 scored 9.6% and Claude Opus 5.5 scored 9.2%. Sonnet 5.5 had the highest single-attempt pass rate of the three.
Written by AI from the coverage · updated 2 hours ago
Coverage
Follow the reports to see the story from different sides.
- Hacker News front pageDayjob: Tax – evaluating AI agents on 50 professional tax assignments
Surge AI 推出 DAYJOB: Tax，涵盖八类税务业务的 50 个任务环境，评测 AI 智能体完成专业税务工作的能力。以满足至少 95% 任务标准为通过条件，Claude Sonnet 5.5 的 pass@1 为 10.8%，Claude Fable 5.1 和 Claude Opus 5.5 分别为 9.6% 和 9.2%。
Heat over time
Not enough continuous observations to draw a trend yet.