About Cloudflare AI Gateway
Cloudflare AI Gateway is a service that sits between your application and AI providers. It adds caching, rate limits, analytics and fallbacks without changes to the model code.
What it does
AI Gateway receives requests from your app and forwards them to the provider you choose. Responses to repeated requests can be cached, and rate limits can cap how often calls are made. If one model fails, the gateway can fall back to another. Logs, analytics and cost views show how AI traffic is used. Many providers are supported through one endpoint, including a unified API that follows the OpenAI format.
Key features
- Cache responses to repeated AI requests
- Rate limits and spend limits for each gateway
- Model fallback when one provider fails
- Analytics, logs and cost views for AI traffic
- One endpoint for many providers
Who it's for
The service suits developers and platform teams who already use Cloudflare or who want a hosted proxy for AI calls. Teams that call one model directly may not need a proxy.
Pricing and good to know
The documentation says the gateway is available on all Cloudflare plans. Specific prices and usage limits are on the pricing page. Some features, such as spend limits, are marked as beta, so check their status before you rely on them in production.
Highlights
- Cache responses to repeated AI requests
- Rate limits and spend limits for each gateway
- Model fallback when one provider fails
- Analytics, logs and cost views for AI traffic
- One endpoint for many providers
