Loading leaderboard data…
AA-Briefcase v1.1: Agentic Knowledge Work Benchmark
As of 29 Oct 2026 · 242 ranked rows · Original source ↗
Cost per Intelligence Index task uses Artificial Analysis profile values captured 29 Oct 2026, weighted across the Index evaluations. Workbook cost values mentioned in the source notes below are evaluation-specific.
The source places Analytical Quality and Presentation in one tab with separate graph indicators. Open each complete ranking below.
Elo from pairwise comparisons of professional presentation. This is a separate indicator within the source's shared Analytical Quality & Presentation Elo tab.
1 results
| 45 | OpenAI | GPT-6 Astra (Medium) | 1417 | -13.96 / +14.98 | Sep 2026 | $2.73 |