Deepnote Launches Agent Initiative Benchmark; $0.11 Model Beats Priciest Rivals

Deepnote Research released the Agent Initiative Benchmark, open-sourcing a fictional company PageMantle with six years of detailed financial, CRM and billing data to test 12 frontier models' unprompted proactive help. A $0.11 model outperformed the three most expensive models.

2026-11-13 ~ 2026-11-13 · 3 related posts

• OpenAI's $0.11 model beats the priciest LLMs in new Slack agent benchmark — heyshrutimishra · 2026-11-13
• Deepnote open-sources synthetic company to benchmark proactive agents across 12 frontier models — heyshrutimishra · 2026-11-13

1 near-duplicate retellings: alex_verem