Databricks is rolling out GPT‑5.5 to its enterprise customers after the model crushed its internal benchmark, OfficeQA Pro. This test is no joke—it throws scanned PDFs, ancient legacy files, and long-context docs at models, tasks that usually trip up production agent systems. GPT‑5.5 hit 50% accuracy (a new state of the art) and slashed errors by 46% compared to GPT‑5.4, especially in parsing gnarly old documents where a single missed digit can derail an entire workflow. The model also got smarter at orchestrating multi-step tasks, avoiding those pointless search detours that wasted compute with the previous version. Now, through Databricks’ AI Unity Gateway, companies can plug GPT‑5.5 into custom agent workflows built with AgentBricks and the Agent Supervisor API. Why it matters: This signals that frontier models are finally getting reliable enough to handle the messy, real-world document chaos that defines most enterprise use cases.