Skip to content
MyStackPlan

Cloudflare AI Gateway pricing

A proxy in front of your AI model calls that adds caching, rate limits and analytics. Core features are free. Paying through Cloudflare adds a 5% credit fee.

Every plan below was checked against Cloudflare AI Gateway's own pages on . Prices in USD.

Plans

Core (your own provider keys)

Free

At the limit
Stops at the limit
  • Core features are free on all plans: dashboard analytics, caching and rate limiting. You need a Cloudflare account.
  • You pay each AI provider directly with your own key, so there is no gateway fee on tokens.
  • Limit of 10 gateways per account on the free plan and 20 on a paid plan.
  • Logs: accounts that created their first gateway on or after September 24, 2026 store logs in Workers Logs and follow its pricing. Older accounts use Legacy Logs, capped at 100,000 logs on the Workers Free plan and 10 million per gateway on Workers Paid.
  • Guardrails are billed as Workers AI inference. DLP scanning is free. Logpush exports are billed at Logpush rates.

Source: developers.cloudflare.com · checked

Unified Billing (pay through Cloudflare)

Pay as you go

At the limit
Bills for extra usage
Usage pricing
  • AI model spend: $0.05 per USD
  • Instead of keeping separate provider accounts, you buy credits from Cloudflare. A 5% fee applies to credits purchased, so a $100 purchase is charged as $105.
  • Provider token prices are passed through with no markup.
  • Requests using Cloudflare-managed credentials are limited to 200 requests per 60 seconds per gateway, with 429 errors above that. The limit does not apply when you use your own provider keys.

Source: developers.cloudflare.com · checked

Works well with

  • Anthropic Claude APICloudflare AI Gateway can sit in front of Anthropic requests to add logging, caching and rate limits.
  • Cloudflare WorkersWorkers can call AI models through AI Gateway using an AI binding.
  • Google Gemini APICloudflare AI Gateway can sit in front of Google requests to add logging, caching and rate limits.
  • OpenAI APICloudflare AI Gateway can sit in front of OpenAI requests to add logging, caching and rate limits.

Alternatives for ai gateways & sandboxes