--- title: "AI Gateway 2.x concepts" description: This page describes the differences between the AI Gateway running on Kong Gateway plugin model (V1) and the new AI Gateway 2.x Policies model. url: "/ai-gateway/ai-gateway-v2-concepts/" canonical_url: "/ai-gateway/ai-gateway-v2-concepts/" content_type: reference min_version: ai-gateway: '2.0' products: - AI Gateway tags: - ai canonical: true works_on: - konnect --- # AI Gateway 2.x concepts AI Gateway 2.x introduces a dedicated control plane for AI workloads in Konnect. Instead of requiring users to manually build AI behavior on top of Kong Gateway through proxy plugins, AI Gateway exposes first-class AI entities: Providers, Models, MCP Servers, and Agents. This guide explains what changed, maps each AI entity to its corresponding proxy plugin configuration, and walks you through migrating an existing configuration using the `kongctl` AI Gateway conversion extension. This guide is intended for teams running AI Gateway on Kong Gateway 3.x who want to move to the AI Gateway 2.x control plane. If you are starting fresh, see [Get started with AI Gateway](/ai-gateway/get-started/). ## What's changing In AI Gateway running on Kong Gateway, AI functionality is delivered by three proxy plugins that extend Kong Gateway's core proxying. You build Services and Routes, then attach a plugin to add AI behavior: - AI Proxy Advanced provides model proxying, transformation, and load balancing across providers and models. - AI MCP Proxy bridges Kong-managed Services to the Model Context Protocol, converting REST APIs into MCP tools or fronting upstream MCP servers. - AI A2A Proxy adds observability and gateway control for Agent-to-Agent protocol traffic. This model works, but it couples every AI concept to Kong Gateway primitives. A single logical model can require a Service, a Route, an AI Proxy Advanced plugin, and several supporting plugins, with the AI intent spread across all of them. AI Gateway 2.x abstracts those plugins into a purpose-built entity model on its own control plane. You no longer need to configure Services, Routes, and plugins manually. Instead, you declare the AI resource you want, and the control plane provisions the underlying primitives for you. ### Entity mapping The following table describes how the two models relate at a high level: #### [AI Proxy Advanced](/plugins/ai-proxy-advanced/) on a Service or Route V2 (Native AI Gateway model): [AI Model](/ai-gateway/entities/ai-model/) Description: One model entry per virtual model, with one or more `targets`. #### Set `config.targets[].model.provider` on [AI Proxy Advanced](/plugins/ai-proxy-advanced/) with inline auth V2 (Native AI Gateway model): [AI Model Provider](/ai-gateway/entities/ai-model-provider) Description: Provider credentials are now declared once and reused across AI Model entities. #### Set `config.targets[].route_type` on [AI Proxy Advanced](/plugins/ai-proxy-advanced/) V2 (Native AI Gateway model): Set `capabilities` and `formats.type` on an [AI Model](/ai-gateway/entities/ai-model/) Description: The `route_type` is decomposed into a `capabilities` array and a format `type`. #### Set the `config.model.options[].input_cost` and `config.model.options[].output_cost` on [AI Proxy Advanced](/plugins/ai-proxy-advanced/) V2 (Native AI Gateway model): Set [Model cost parameters](/ai-gateway/model-cost-management/#model-cost-configuration) on an [AI Model](/ai-gateway/entities/ai-model/) Description: Additional cost management features are available to match how LLM providers actually bill. #### Set `config.balancer` on [AI Proxy Advanced](/plugins/ai-proxy-advanced/) V2 (Native AI Gateway model): Set `config.balancer` on an [AI Model](/ai-gateway/entities/ai-model/) Description: The same load balancing algorithms are available. #### Set `config.vectordb` and `config.embeddings` on [AI Proxy Advanced](/plugins/ai-proxy-advanced/) V2 (Native AI Gateway model): Set `config.balancer.AIGatewayModelBalancerSemanticConfig.vectordb` and `config.balancer.AIGatewayModelBalancerSemanticConfig.embeddings` on an [AI Model](/ai-gateway/entities/ai-model/) Description: Carried over with the same Redis and pgvector strategies. #### [AI MCP Proxy](/plugins/ai-mcp-proxy/) on a Service or Route V2 (Native AI Gateway model): [AI MCP Server](/ai-gateway/entities/ai-mcp-server/) Description: Each plugin `mode` maps directly to an AI MCP Server `type` value in version 2.x. Additionally, a new `upstream-server` type is available. #### Set `config.default_acl` and `config.tools.acl` on [AI MCP Proxy](/plugins/ai-mcp-proxy/) V2 (Native AI Gateway model): Set `access` or `tools.access` on an [AI MCP Server](/ai-gateway/entities/ai-mcp-server/). Configure an [AI Consumer](/ai-gateway/entities/ai-consumer/) or [AI Consumer Group](/ai-gateway/entities/ai-consumer-group/) Description: ACLs become first-class fields. #### [AI A2A Proxy](/plugins/ai-a2a-proxy/) on a Service or Route V2 (Native AI Gateway model): [AI Agent](/ai-gateway/entities/ai-agent/) Description: First class A2A support with URL rewriting and A2A analytics built in. #### [Plugins](/plugins/) V2 (Native AI Gateway model): [Policies](/ai-gateway/policies/) Description: AI Policies replace plugins, and can be attached to other entities. The `type` field on a Policy corresponds to the plugin. #### Consumers and Consumer Groups V2 (Native AI Gateway model): [AI Consumer](/ai-gateway/entities/ai-consumer/) and [AI Consumer Group](/ai-gateway/entities/ai-consumer-group/) Description: Managed from the control plane. #### Vault V2 (Native AI Gateway model): [AI Vault](/ai-gateway/entities/ai-vault/) and [AI Data Plane Certificates](/ai-gateway/entities/ai-data-plane-certificate/) Description: Referenceable fields keep the same {vault://...} syntax. Note the following terminology changes: - AI Policies replace Kong Gateway plugins. All AI Policies have some common parameter. Each AI Policy has a `type` which corresponds to a plugin from AI Gateway running on Kong Gateway, such as `ai-sanitizer` or `openid-connect`, and their `config` is the same as the plugin. - AI Model Providers are now separate reusable entities. This decouples config and credentials of upstream providers from specific models, which allows you to declare an AI Model Provider once and reference it by name from multiple AI Models. - A Route from AI Gateway running on Kong Gateway is split into two AI Gateway 2.x concepts: a `capabilities` list and a `formats` entry. ## Related Resources - [Migrate to AI Gateway 2.x](/ai-gateway/v2-migration-guide/) - [AI Gateway Policies](/ai-gateway/entities/ai-policy/) - [AI Gateway entities](/ai-gateway/entities/) - [Using kongctl to manage AI Gateway](/ai-gateway/kongctl/) - [AI Gateway architecture](/ai-gateway/architecture/)