Databricks brings GPT-5.5 to enterprise agent workflows
OpenAI
Databricks has integrated GPT-5.5 into its enterprise agent workflows, achieving record-breaking accuracy on the OfficeQA Pro benchmark for complex document tasks.Summary
Databricks has officially integrated GPT-5.5 into its enterprise workflows, citing its new state-of-the-art performance on the OfficeQA Pro benchmark. The model demonstrated a 46% reduction in errors compared to its predecessor, GPT-5.4, particularly excelling in parsing scanned PDFs and legacy enterprise documents. By utilizing the AI Unity Gateway alongside AgentBricks and the Agent Supervisor API, enterprise customers can now leverage GPT-5.5 to improve multi-step task orchestration, context retrieval, and overall agent reliability without the need for constant supervision.
(Source：OpenAI)