AI Infrastructure

AI Gateway Comparison 2026: Kong, LiteLLM, Portkey, Cloudflare, Vercel, Azure, AWS and Self-Hosted Options

An LLM gateway comparison for platform and security teams: nine gateways checked against their own documentation for routing, limits, caching, guardrails, MCP support, licence and where they can run. Verified September 2026.

An AI gateway, also called an LLM gateway, is the proxy between applications and model providers that routes each request, fails over when a provider breaks, enforces token and spend limits and logs usage. Comparing gateways comes down to where each one runs, what it enforces before a call is made, and whether it also governs MCP tool traffic.

What to compare in an AI gateway

Most gateways now share a baseline: an OpenAI-compatible endpoint, many providers behind it, retries and fallbacks, and request logs with token counts. The differences that decide a purchase sit elsewhere.

  • Where it runs. On the vendor’s network, in your own cloud account, or on-premises with no outside dependency. This settles whether prompts ever leave your perimeter.
  • What it enforces before the call. Token and spend limits, model allowlists, content guardrails and PII redaction only count as controls if they run in the request path.
  • Licence and tiers. Several open-source gateways keep specific features, such as semantic caching or cost-based rate limiting, in paid editions.
  • Tool traffic. Agents call tools as well as models. Some gateways now proxy or generate MCP servers; others handle model calls only.
  • Evidence. What a log records per call, and whether it is enough to reconstruct why a given model answered.

If you are new to the category, start with what an on-premises AI gateway does and come back for the comparison.

AI gateways compared at a glance

Each entry below reflects the vendor’s own documentation or repository, verified September 2026. A dash means the feature was not documented on the pages we checked, not that it is absent.

Where each gateway runs and how it is licensed

GatewayHostingSource and licenceMCP and tool traffic
Kong AI GatewaySelf-hosted Kong Gateway, or Kong Konnect (managed)Kong Gateway is Apache 2.0; key AI plugins need AI Gateway EnterpriseAI MCP Proxy: REST-to-MCP conversion and MCP proxying with per-tool ACLs (Enterprise)
LiteLLMSelf-hosted (Docker, Terraform for AWS and GCP) or hostedMIT; enterprise features such as SSO under a commercial licenceMCP gateway
Portkey AI GatewaySelf-hosted (Docker, Node.js, Cloudflare Workers), Portkey Cloud, or enterprise private cloudMIT; some features hosted or enterprise onlyMCP gateway with authentication and access control
Agent Router (formerly Envoy AI Gateway)Kubernetes, local, or hosted by TetrateApache 2.0One filtered MCP tool catalogue per agent
TrueFoundry AI GatewaySaaS, VPC, on-premises or air-gappedVendor productMCP gateway
Cloudflare AI GatewayManaged, on Cloudflare’s networkManaged service–
Vercel AI GatewayManaged, callable from any infrastructureManaged service–
Azure API Management AI gatewayAzure-managed; capabilities vary by tierManaged serviceExposes REST APIs as MCP servers; passthrough to existing MCP servers
Amazon Bedrock AgentCore GatewayFully managed on AWSManaged serviceConverts APIs, Lambda functions and services into MCP tools
AWS Multi-Provider Generative AI GatewayDeployed in your AWS account on ECS or EKSReference architecture built on LiteLLMAs LiteLLM

What each gateway enforces

GatewayRouting and fallbackLimits and spendCachingGuardrails
Kong AI GatewayLoad balancing and failover across providersToken- or cost-based limits per consumer, model or provider (Enterprise)Semantic cache (Enterprise)Prompt guards, PII sanitiser, AWS, Azure and GCP guardrail integrations
LiteLLMRetries, fallbacks, load balancingSpend tracking, budgets, rate limits, virtual keysYesYes
Portkey AI GatewayRetries, fallbacks, load balancing–Simple (open source); semantic (hosted and enterprise)50+ guardrails
Agent RouterProvider fallback, model name virtualisationToken limits per team, app or model––
TrueFoundry AI GatewayLatency-based routing and fallbackCost- or token-based quotas–PII filtering, toxicity detection
Cloudflare AI GatewayRetries, model fallback, dynamic routingRate limits, spend limits, cost analyticsIdentical-request cachingGuardrails, data loss prevention
Vercel AI GatewayProvider ordering, provider and model fallbacksBudgets per team, project, key or memberConfigurable–
Azure API Management AI gatewayWeighted and priority load balancing, circuit breakerToken limits and quotas per consumerSemantic caching with Azure Managed RedisAzure AI Content Safety policy
Amazon Bedrock AgentCore GatewayModel-based routing across providers–––
AWS Multi-Provider Generative AI GatewayVia LiteLLMVia LiteLLMPrompt caching with ElastiCacheAmazon Bedrock guardrails

Open-source and self-hostable gateways

Kong AI Gateway

Kong extends its API gateway to model and tool traffic. Its documentation describes one consistent API across providers including OpenAI, Anthropic, Azure AI, Amazon Bedrock and Gemini, and the AI Proxy plugin also lists self-hosted backends such as Ollama and vLLM. The Kong Gateway repository is Apache 2.0 and calls itself an API, LLM and MCP gateway. The plugins an enterprise usually wants most are marked “AI Gateway Enterprise” on Kong’s plugin pages: AI Semantic Cache, AI Rate Limiting Advanced, which limits by total, prompt or completion tokens or by cost, and AI MCP Proxy, which needs Gateway 3.12 or later and converts REST APIs into MCP tools with per-tool ACLs. Guardrail plugins include prompt guards, a semantic prompt guard, a PII sanitiser and integrations with AWS Guardrails, Azure Content Safety, GCP Model Armor and Lakera. It runs self-hosted or in Kong’s managed Konnect platform.

LiteLLM

LiteLLM is a Python SDK plus a proxy server that exposes more than 100 LLM APIs in the OpenAI format. The proxy handles retries, fallbacks and load balancing, virtual keys, spend tracking per project and user, budgets and rate limits, guardrails, caching and an MCP gateway, with logging callbacks to tools such as Langfuse and MLflow. The core is MIT licensed, and enterprise features such as SSO sit under a separate commercial licence. It runs from Docker, from Terraform modules for AWS and GCP, or as a hosted service. AWS’s own multi-provider gateway guidance is built on it.

Portkey AI Gateway

Portkey’s open-source gateway is MIT licensed and routes to more than 1,600 models across 45+ providers, including Ollama for local models. It offers automatic retries of up to five attempts, fallbacks, load balancing, more than 50 guardrails, role-based access control and key management. Simple caching is in the open-source edition; semantic caching is listed for the hosted and enterprise versions. It also describes an MCP gateway with authentication, access control and observability. You can self-host with Docker, Node.js or Cloudflare Workers, use Portkey Cloud, or buy the enterprise edition for private cloud on AWS, Azure, GCP, OpenShift or Kubernetes.

Agent Router, formerly Envoy AI Gateway

Envoy AI Gateway has been renamed Agent Router and is now an Agentic AI Foundation project. The project says the code and maintainers are unchanged, and container images, Helm charts, CRDs and the aigw CLI keep their names. It runs on Envoy Proxy and Envoy Gateway and offers one OpenAI-compatible API for hosted providers, self-hosted inference and MCP servers. Features include provider fallback, model name virtualisation, token limits per team, app or model, centralised credentials, OpenTelemetry telemetry and a single MCP tool catalogue filtered per agent. It is Apache 2.0 and runs locally, on Kubernetes, or as Tetrate’s hosted Agent Router Service. Teams that already operate Envoy will find it the most familiar option.

TrueFoundry AI Gateway

TrueFoundry sells its gateway as part of a broader platform. The product page lists access to more than 1,600 models, latency-based routing with fallback, cost- or token-based quotas scoped with metadata filters, guardrails including PII filtering and toxicity detection, an MCP gateway and request-level observability. It is one of the few commercial gateways whose page names SaaS, VPC, on-premises and air-gapped deployment.

Managed gateways from cloud and platform providers

Cloudflare AI Gateway

Cloudflare runs its gateway on its own network. The features page lists caching of identical requests, rate limiting, spend limits by model, provider or custom metadata, dynamic routing by segment, geography or A/B test, guardrails for harmful content, data loss prevention scanning for PII and financial data, token-based authentication, stored provider keys, analytics and logging, and custom costs for negotiated rates. Request retry and model fallback are also documented. It suits teams already on Cloudflare; prompts and responses pass through Cloudflare’s infrastructure.

Vercel AI Gateway

Vercel’s gateway is a managed service that any server, cloud or local environment can call with an API key; Vercel deployments can use OIDC instead. It accepts the AI SDK, OpenAI Chat Completions and Responses, and Anthropic Messages formats. It routes across providers and fallback models and records status, provider, latency, token usage, cost and every routing attempt. Budgets can be set per team, project, API key or member, but the documentation, updated 14 September 2026, notes that budgets cover spend billed through Vercel’s system credentials, that bring-your-own-key spend is metered separately, and that budgets behave as soft caps.

Azure API Management AI gateway

In Azure the AI gateway is a set of capabilities in API Management rather than a separate product, and Microsoft notes that availability varies by service tier. It governs OpenAI Chat Completions and Responses, Anthropic Messages on v2 tiers and Google Vertex AI APIs, with a unified OpenAI-compatible model API in preview. The llm-token-limit policy sets tokens-per-minute limits or quotas per consumer, semantic caching uses Azure Managed Redis, and a content safety policy calls Azure AI Content Safety. Backends support round-robin, weighted, priority and session-aware load balancing with a circuit breaker. Prompts and completions can be logged to Azure Monitor. On the tool side it can expose existing REST APIs as MCP servers, pass traffic through to MCP servers and manage A2A agent APIs.

AWS options

AWS offers two different routes. Amazon Bedrock AgentCore Gateway is a fully managed entry point for agent traffic: it converts APIs, Lambda functions and services into MCP tools, fronts other agents, and routes inference across model providers through a model-based routing endpoint, with inbound and outbound authentication and semantic tool search. The Guidance for Multi-Provider Generative AI Gateway on AWS is a reference architecture you deploy yourself: LiteLLM on ECS or EKS behind AWS WAF and a load balancer, with ElastiCache for settings and prompt caching, RDS for virtual keys, Secrets Manager for provider credentials, Bedrock for guardrails and logs written to S3.

Self-hosted and on-premises versus managed gateways

This is the decision that narrows the list fastest, and it turns on the data path more than on features.

ConsiderationManaged gatewaySelf-hosted or on-premises gateway
Prompt and response pathThrough the vendor’s networkStays on infrastructure you run
Air-gapped operationNot possiblePossible when every model behind it is local
Patching and scalingDone by the vendorDone by your platform team
Provider credentialsHeld by the vendor unless you bring your own keysHeld in your own secret store
Logs and auditVendor console and exportsStraight into your SIEM and retention policy
Feature accessUsually the full feature setSome features gated behind enterprise tiers

Two points are easy to miss. A self-hosted gateway is not automatically private: if every model behind it is a public API, prompts still leave the building, so gateway placement only pays off alongside local models or approved private endpoints. And logs are sensitive data in their own right, because a gateway that stores prompts and completions holds a copy of everything users asked. Decide retention before you switch logging on. Our guide to air-gapped deployments covers what a fully disconnected setup requires.

How to choose an LLM gateway

  1. Fix the data path first. Decide which workloads may send prompts to a third-party network and which may not. That single decision removes half the options for regulated teams.
  2. List controls that must be enforced, not observed. Token limits, spend caps, model allowlists and content checks should block a request before it reaches a model. A dashboard that shows an overspend afterwards is reporting, not control.
  3. Read the tier labels. For each open-source candidate, confirm whether the features you need, such as semantic caching, cost-based limits, SSO or MCP support, are in the free edition.
  4. Decide whether the gateway also governs tools. If agents will call MCP servers, choose between a gateway that handles both kinds of traffic and a separate tool registry with its own grants and audit.
  5. Look at how routing decisions are made. Static rule tables and provider ordering are easy to reason about but drift as models and prices change. Our explainer on how LLM routing works and the cost, latency and quality formula set out the trade-offs.
  6. Be deliberate about caching. Semantic caching saves money and can serve stale or cross-tenant answers; see the semantic caching trade-offs before enabling it for regulated workloads.
  7. Model the spend. Estimate what routing, caching and limits would save on your actual traffic with the AI savings calculator before committing to a licence.

How VDF AI fits

VDF AI’s gateway is VDF AI Router, a Docker-packaged service that runs on your VMs, Kubernetes or bare metal and exposes a REST API and a Python SDK. It reaches local models through Ollama and cloud models through OpenRouter. Policy is the first layer of every decision: regulated domains are restricted to approved models, workloads can be pinned to a named model, and organisation-wide allow and deny lists apply before any call. Air-gap mode disables external APIs entirely. Each decision returns the selected model, a readable reason, up to five ordered failover candidates and per-model scores, and the SEEMR learning engine refines routing from observed quality, latency, failures and energy, with a configurable share of traffic, around two per cent, sent to a challenger model.

Budgets and rate limits sit in the request path. The Router does not publish semantic caching, PII redaction or prompt-injection filtering, so teams that need those should add dedicated controls or pick a gateway above that documents them. Tool traffic is handled by a separate component: the MCP gateway grants tools per role and writes tool calls to the same audit trail as routing decisions. It is a commercial product, and on-premises deployments are licensed as an annual capacity pool.

Sources

Frequently asked questions

Which AI gateways can run on-premises or in an air-gapped network?

Self-hostable gateways are the candidates: Kong Gateway, LiteLLM, the open-source Portkey gateway and Agent Router, formerly Envoy AI Gateway, can all run on infrastructure you operate, and TrueFoundry lists on-premises and air-gapped deployment. An air-gapped setup also needs every model behind the gateway to run locally, because a self-hosted gateway that forwards to public APIs still sends prompts out. Managed services such as Cloudflare AI Gateway and Vercel AI Gateway run on the vendor's network.

Which AI gateways support MCP tool traffic?

According to their documentation in September 2026, Kong offers an AI MCP Proxy plugin in its enterprise tier, LiteLLM and Portkey include MCP gateway features, Agent Router consolidates MCP servers into one filtered catalogue per agent, TrueFoundry lists an MCP gateway, Azure API Management can expose REST APIs as MCP servers and pass traffic through to existing ones, and Amazon Bedrock AgentCore Gateway converts APIs and Lambda functions into MCP tools. Check whether MCP support sits in the free or the paid tier.

Is an open-source LLM gateway enough for enterprise use?

Often it is a good core, but check three things. First, which features the open-source edition actually includes, because several projects keep semantic caching, advanced rate limiting, SSO or MCP features in paid tiers. Second, who will run, patch and scale it, since a gateway sits on the critical path of every AI feature. Third, whether its logs give auditors what they need, such as which model answered, under which policy and at what cost.

Should an AI gateway cache LLM responses?

Caching can cut latency and cost for repeated questions, but it needs care. Exact-match caching is predictable. Semantic caching returns a stored answer for a prompt that is merely similar, which can serve stale or wrong content and, in multi-tenant settings, can leak one user's answer to another if cache keys ignore permissions. Enable it per workload, scope cache keys to tenant and access rights, set expiry, and exclude regulated or personalised requests.

What should we test during an AI gateway proof of concept?

Test failure and enforcement rather than the happy path. Take one provider offline and watch the fallback order. Push a workload past its token or spend limit and confirm the gateway blocks it before the call. Send a request that policy forbids from reaching an external model. Then export a day of logs and check that each entry shows the model used, the reason, the cost and the caller, in a format your security team can ingest.

Filed under
AI gatewayLLM gatewaymodel routingMCP gatewayAI infrastructureon-premises AI
AI Agent Platform

Evaluating AI agent platforms?

See what a governed agent platform looks like in production — build, orchestration, retrieval, routing and audit in one product — then score it against the platforms on your shortlist.

Or start free — no credit card →

Keep reading