The new standalone platform provides unified MCP tool governance, modality-aware cost management, identity-aware AI policies, and expanded provider support
, /PRNewswire/ — Kong Inc., the AI Connectivity Company, today announced the general availability of Kong AI Gateway 2.2, featuring a new AI-native user experience, advanced cost management, and broader model support. With 2.2, Kong AI Gateway brings the same governance, security, and operational discipline enterprises expect from API management to the fast-moving world of agentic AI.
AI is outpacing traditional infrastructure. New models, agent architectures, protocols, and pricing models are emerging in weeks rather than quarters, creating a growing gap between the pace of AI innovation and the infrastructure enterprises rely on to govern it. As AI connectivity spans models, agents, intelligent tools, and context across increasingly diverse environments, governance can no longer be limited to a single part of the stack. Kong AI Gateway closes that gap by giving AI its own platform and release cadence, so enterprises can adopt what’s next in AI without compromising security, compliance cost control, resilience, or visibility.
Kong AI Gateway 2.2 includes a wave of new capabilities, including:
- Governed MCP tool access through a single endpoint: MCP Server Bundling consolidates a growing sprawl of MCP servers into one Kong route, revealing only the tools each caller is authorized to see and execute.
- Cost governance that matches how AI is actually priced: Dynamic, modality-aware pricing tracks text, audio, image, and video separately, along with cache reads versus cache writes, giving platform and finance teams accurate cost attribution instead of flat, token-based estimates. Kong AI Gateway 2.2 also ships support for Headroom for advanced prompt compression for cost optimization.
- Identity-aware AI policies: Powered by Kong Identity, rate limiting, analytics, cost attribution, and access control can now key on an authenticated principal rather than a gateway-local Consumer, giving enterprises a consistent way to know who, or what agent, is behind every call.
- Native AWS IAM authentication for Amazon Bedrock AgentCore: Declarative, platform-managed SigV4 authentication replaces hand-rolled workarounds for AWS-native teams running agents and MCP servers on AgentCore.
- Expanded model and provider coverage: Native support added for Kimi, Microsoft Foundry, and Amazon SageMaker, giving enterprises more flexibility in where and how they run AI workloads without sacrificing governance.
- Support for Jev: Developers can adopt this emerging decision-oriented model without creating a separate integration and governance path for a new class of AI workload. Jev becomes another choice developers can make while keeping the infrastructure around that choice consistent.
“The latest update to AI Gateway is one of the most exciting changes we have made to date. The promise of agentic AI is not just about what an agent can do, but whether an enterprise can confidently put it to work. That requires visibility into what is happening, control over what agents can access and execute, and the ability to manage the risks and economics that come with it. That’s the foundation of AI governance we’re giving customers with AI Gateway 2.2, and it’s what will make agentic AI viable in the enterprise,” said Reza Shafii, Senior Vice President, Product, Kong Inc.
Now generally available in Kong Konnect, the AI Connectivity Platform, with no beta enablement required. To learn more visit https://konghq.com/products/kong-ai-gateway.
About Kong Inc.
Kong Inc., the AI Connectivity Company, is building the connectivity layer of AI. Trusted by the Fortune 500® and AI-native startups alike, Kong’s unified API and AI platform enables organizations to secure, manage, accelerate, govern, and monetize the flow of intelligence across APIs and AI traffic — on any model, any cloud. For more information, visit www.konghq.com.
SOURCE Kong Inc.
