Databricks' suite for building, governing, and serving production AI — agents, RAG, fine-tuning, and evaluation — on top of your governed data.
AI Infrastructure comparison
Pricing, pros, cons, and ideal use cases — side by side.
Databricks' suite for building, governing, and serving production AI — agents, RAG, fine-tuning, and evaluation — on top of your governed data.
An AI gateway that adds routing, caching, observability, and guardrails to LLM traffic through a single control plane.
| Databricks Mosaic AI | Prisma AIRS AI Gateway (formerly Portkey) | |
|---|---|---|
| Pricing | EnterpriseDatabricks now sells its agent tooling as Agent Bricks, billed in DBUs; the rates below are for Premium on AWS US East. Model Serving CPU costs $0.07/DBU, with standard workloads billed at 4 DBU per hour. Databricks Apps, which Agent Runtime runs on, bill App Capacity at $0.75/DBU, and the default Medium app size uses 0.5 DBU per hour. Databricks publishes no separate Agent Runtime price, so confirm whether it is billed as App Capacity. Proprietary-model pay-per-token costs $0.07/DBU. The Knowledge Assistant ($0.15 per answer) and Supervisor Agent are no longer available to new customers. As read on the vendor's pricing page on 10 November 2026. | FreemiumFree developer tier. Paid Pro and Enterprise plans add volume, self-hosting, and governance. |
| Category | AI Infrastructure | AI Infrastructure |
| Ideal for | Enterprises building AI on the Databricks lakehouseData teams that want AI governed alongside dataOrganizations unifying data and AI on one platform | Enterprises standardizing LLM access across teamsPlatform teams needing routing and failoverOrgs wanting cost and reliability controls in one place |
Prisma AIRS AI Gateway (formerly Portkey) is the lighter-weight option (Freemium), while Databricks Mosaic AI sits higher on the pricing ladder (Enterprise). Databricks Mosaic AI is built around enterprises building ai on the databricks lakehouse; Prisma AIRS AI Gateway (formerly Portkey) leans more toward enterprises standardizing llm access across teams. Shortlist the one whose strengths line up with your biggest constraint.
Get one AI workflow a week showing AI Infrastructure in a real stack — what they cost, and where each one breaks.