Databricks deploys GPT-6 Astra: 60% cost increase, gains only on complex tasks

Databricks co-founder Patrick Wendell shared five observations from deploying GPT-6 Astra across the company. Model capability gains are concentrated on high-complexity tasks (system design, long-horizon tasks), while simple tasks show no improvement. Overall coding expenditure rose 60% due to increased usage. The rollout used a gateway-based experiment pipeline with budget-based routing to guide model selection.