Surge AI Launches DAYJOB: Tax Agent Benchmark

Surge AI added a tax benchmark to its DAYJOB family: 50 environments across eight tax practice areas, testing agents' professional judgment alongside calculations. In one example, an agent picked the right sale date but missed a residence deadline affecting a $250,000 exclusion in all five attempts; Opus caught it every time.

• Claude Code status line burns 15-20% of tokens2026-10-29
• Open-weight models edge past Sonnet 5 on some benchmarks2026-10-16
• LegalOn cuts daily Codex costs by 65%2026-10-09
• Blogger says self-hosting AI saves no money but sticks with it2026-10-03
• OpenAI introduces GPT-6.1 Sol2026-09-29