Prisma AIRS AI Gateway

What Prisma AIRS AI Gateway is

Prisma AIRS AI Gateway centralizes control, security, and observability for AI requests that teams and applications send to large language models. It connects to more than 1,600 LLMs across 50 or more providers, including OpenAI, Anthropic, Google, AWS Bedrock, Azure OpenAI, Mistral, Cohere, and self-hosted models. Developers send requests using the standard OpenAI API format, and can switch providers or fallback models without changing application code. Workspaces let teams govern model access, budgets, rate limits, and guardrail policies. Scoped virtual keys control which models users can access and how much they can spend. Inline protections can block prompt injection, identify sensitive data, and apply topic restrictions to both requests and responses. Integrations cover LLM providers, MCP servers, secret references, and agent backends. Credentials are encrypted at rest and decrypted in memory during requests, while raw API keys remain hidden from workspaces. The gateway is available as Palo Alto Networks-hosted SaaS or as Hybrid deployment on a customer-managed Kubernetes cluster; both are managed through Strata Cloud Manager. In Hybrid mode, request and response payloads are processed within the customer's environment and are not sent to Palo Alto Networks. Hybrid deployment requires a provisioned Kubernetes cluster and Helm 3. Licensing is consumption-based, with credit requirements calculated from anticipated token usage and tiered volume discounts; pricing is on request.

Who it suits

Prisma AIRS AI Gateway suits teams and application developers that need to route requests across multiple LLM providers while governing access, budgets, rate limits, and runtime protections. Its Hybrid option is suited to organizations with a provisioned Kubernetes cluster and Helm 3 that want request and response payloads processed in their own environment.

Limits

Pricing is consumption-based, with credits tied to anticipated token usage and requirements calculated by the sales team. Hybrid deployment requires a provisioned Kubernetes cluster and Helm 3.

Getting started

1. Open the Prisma AIRS AI Gateway website.
2. Choose hosted SaaS or Hybrid deployment.
3. For Hybrid, provision a Kubernetes cluster and Helm 3.
4. Manage the gateway through Strata Cloud Manager.
5. Configure provider integrations and workspaces, then send requests using the standard OpenAI API format.
6. Contact a sales representative to determine credit requirements based on anticipated token usage.

Verdict

Choose Prisma AIRS AI Gateway if your teams need a common API path to many LLM providers with centralized budgets, access controls, and inline guardrails. Hybrid deployment may suit organizations that want payload processing within their environment and can provide the required Kubernetes cluster and Helm 3. Look elsewhere if you need a published fixed price or cannot meet the Hybrid deployment requirements.