AI testing agents that write, run and self-heal Playwright end-to-end and API tests, delivered to your repo as pull requests.
Contact sales
Kong AI Gateway
Gateway and control layer for LLM, MCP and agent traffic, with usage metering, billing and cost limits built in.
What is Kong AI Gateway?
Kong AI Gateway is part of the Kong Konnect platform and sits between your applications, AI agents and the models or tools they call. It manages LLM, MCP and A2A traffic through one gateway, applies access control and token-based rate limits, and records which model, customer and request consumed what. A developer portal and a metering and billing module let teams package AI and API access as plans, track usage against entitlements and invoice from the same meters. Gateways can run as Dedicated Cloud Gateways on AWS, GCP or Azure, as serverless or hybrid gateways, or fully self-hosted through the separate Gateway Enterprise offering. Platform and infrastructure teams use it to govern AI spend, add guardrails and turn AI usage into a product. The pricing page lists Plus and Enterprise tiers and shows usage-based rates for control planes, requests and proxied models, so check Kong's pricing page for current numbers.
Use cases
- Route LLM and MCP traffic through one gateway
- Enforce token budgets and rate limits
- Meter and bill AI and API usage
- Monitor model, token and agent consumption
- Publish AI products in a developer portal
- Apply guardrails and prompt controls
- Manage agent identity and access
Who it's for
Why use Kong AI Gateway
- Gives platform teams a single place to see and control AI traffic across models, tools and agents.
- Stops runaway AI spend at the point of request with token-based limits and budgets.
- Lets product teams turn AI and API access into metered, billable plans without building a separate billing pipeline.
- Offers several deployment shapes, from serverless gateways to fully self-hosted, so workloads can run where data must stay.
- Brings agent identity, access control and audit logging into the same platform as the gateway.
Kong AI Gateway features
- Unified gateway for LLM, MCP and A2A traffic
- Token-based rate limits, budgets and quotas at the gateway
- Usage metering down to token, model, customer and request
- Prepaid credits, subscriptions and usage-based billing
- Product catalog with plans, rate cards and add-ons
- Developer portal for AI productization
- Observability for model, token, MCP and agent usage
- OpenTelemetry support
- AI plugins such as semantic cache, prompt guard and sanitizer
- Kong Identity access tokens for agent authentication
- Dedicated Cloud, serverless, hybrid and self-hosted gateway options
- RBAC, SSO and audit logging
Kong AI Gateway integrations
How to get started with Kong AI Gateway
- Choose a gateway type, such as a Dedicated Cloud Gateway, serverless, hybrid or self-hosted, and create a control plane in Konnect.
- Route LLM, MCP and A2A traffic through the gateway and enable AI plugins for guardrails, caching and rate limiting.
- Define features, plans and rate cards in the product catalog and attach meters to the traffic you want to track or bill.
- Publish plans in the developer portal, monitor token and agent consumption, and set quotas and alerts.
- Move to Enterprise if you need SSO, audit logging, a customer success manager or custom limits.
Kong AI Gateway limitations
- Usage-based charges apply for control planes, requests and proxied models
- Fully self-hosted gateways need a separate Gateway Enterprise agreement through sales
- SSO and audit logging are listed as Enterprise only
- Some enterprise AI plugins and the AI Gateway Manager are add-ons unless the plan includes them
Kong AI Gateway pricing
Kong AI Gateway is sold on custom quotes, so its prices aren't published.
Kong AI Gateway plans
Plus
Custom
billed annually
- Unified AI Gateway for LLM, MCP and A2A traffic
- Developer portal and monetization
- Model, token, MCP and agent observability
Enterprise
Custom
- SSO and audit logging
- RBAC
- Dedicated customer success manager
- Professional services
- Custom volume discounts
Plans as listed on the official pricing page on . Prices can change and may vary by country or exclude tax. Check the official pricing page.
Kong AI Gateway alternatives
The main Kong AI Gateway alternatives listed on Debuly are Portkey, LiteLLM, Helicone, Future AGI, nRouter and OrcaRouter. The table compares pricing and platforms; order is A–Z.
| Tool | Pricing | Starting price | Platforms |
|---|---|---|---|
| Kong AI Gateway | Contact sales | — | Web, API, Self-hosted |
| Future AGI | Freemium | — | Web, API, Self-hosted |
| Helicone | Freemium | $79/mo | Web, API, Self-hosted |
| LiteLLM | Open source | — | Python, API, Self-hosted, CLI |
| nRouter | Paid | $5, usage-based | API, Web, Python SDK |
| OrcaRouter | Freemium | $499/mo | API, Web |
| Portkey | Freemium | $49/mo | Web, API, Self-hosted |
Future AGI
Platform for tracing, evaluating, simulating and guarding LLM apps and AI agents, with a gateway, usage-based billing and a free tier.
FreemiumVisit Future AGI (official site)
Helicone
Open-source LLM observability and AI gateway that logs requests, tracks cost and latency, and routes to 100+ models through one API.
FreemiumVisit Helicone (official site)
LiteLLM
Open-source AI gateway and Python SDK that gives one OpenAI-format API to 100+ LLM providers, with keys, budgets and cost tracking.
Open sourceVisit LiteLLM (official site)
nRouter
LLM gateway with one OpenAI-compatible API key, budgets, guardrails and pay-as-you-go credits at list price.
OrcaRouter
OpenAI-compatible AI gateway that routes requests across 200+ models at provider token prices, with guardrails and spend controls.
FreemiumVisit OrcaRouter (official site)
Portkey
AI gateway with observability, guardrails, caching and prompt management for teams running LLM apps in production.
FreemiumVisit Portkey (official site)
Kong AI Gateway FAQ
How much does Kong AI Gateway cost?
Kong's pricing page shows Plus and Enterprise tiers with the price listed as custom, billed annually, plus usage-based rates for items such as gateway control planes, requests and proxied models. Contact Kong for a quote that matches your volume.
Does Kong AI Gateway have a free plan?
The pricing page does not list a free plan. It shows Plus and Enterprise tiers, so check Kong's site or talk to the sales team about trials or evaluation options.
What plans does Kong AI Gateway have?
The pricing page names two tiers, Plus and Enterprise. Plus covers smaller LLM, MCP and A2A deployments, while Enterprise adds SSO, audit logging, RBAC, a customer success manager and professional services.
Can Kong AI Gateway be self-hosted?
Yes. Self-hosted AI gateways with the control plane and data plane both running locally fall under the separate Gateway Enterprise offering, and Kong sets that pricing through its sales team.
Who is Kong AI Gateway for?
It suits platform, API and security teams that need to govern how applications and agents use LLMs and tools, and product teams that want to meter and bill that usage.
Does Kong AI Gateway support observability tools?
The pricing page lists OpenTelemetry support and integration with third-party monitoring and observability solutions, alongside built-in analytics for requests, tokens and agent consumption.
Sources & verification
Facts on this page come from the official sources below, each with the date we read it. Written by Debuly; the vendor did not review this page.
| Source | Type | Accessed |
|---|---|---|
| konghq.com/ | official | |
| konghq.com/pricing | pricing | |
| konghq.com/products/kong-konnect/features/usage-based-metering-and-billing | pricing | |
| developer.konghq.com/metering-and-billing/product-catalog/ | pricing |
Image credits
- Logo: from konghq.com · fetched · Vendor trademark, used only to identify the product; removed on request
- Cover: none collected yet, so a generated tile is shown.
Spot an error? Report an issue with this listing. We also remove a logo or image on the owner's request.
More in Developer platforms 4
All developer platforms toolsConclusion Models and API that give AI agents structured, policy-aligned, traceable decisions.
PaidAI cloud with pay-per-token model APIs for 200+ models, dedicated GPU endpoints, agent sandboxes and GPU rentals.
PaidOpen-source Microsoft framework for end-to-end web testing and browser automation across Chromium, Firefox and WebKit, with CLI and MCP for agents.
Open source
