Back Back
Models6/day by emailPrimary

On Long-Horizon-Terminal-Bench 2.0, 77 tasks across 13 domains with three hours allowed for each, the strongest of 15 frontier models tested, GPT-6-astra, resolved 37 tasks, followed by GPT-5.6-sol with 32, according to an October 2026 arXiv paper

Zongxia Li et al. reported in an October 2026 arXiv paper that GPT-6-astra resolved 37 tasks on Long-Horizon-Terminal-Bench 2.0, followed by GPT-5.6-sol with 32 tasks. This measured the number of tasks resolved by 15 frontier models on Long-Horizon-Terminal-Bench 2.0, which contains 77 tasks across 13 domains with three hours allowed for each.

Verified 2026-10-27

Track with Insider

Why it matters

Hours-long unsupervised work is what agents must master to replace real jobs, and this benchmark reveals current frontier model capabilities.

FREE · NO CARD

Facts like this, six days a week

The Daily Receipts on weekdays, The Weekly Brief on Saturday — six sourced facts a day by email, two more than the free site shows. Every one traced to the filing it came from. Free.

Get the free daily brief

Free · No card · Unsubscribe anytime

Join now and you start with 100 Receipts on The Receipts Ledger — your first step toward a free year of Insider or Pro. How it works →

See today's issue →