AI Gateway - Control AI costs with spend limits
Cloudflare has introduced spend limits for its AI Gateway, a feature that enables cost-based budgets to control AI spending. Unlike rate limiting, which caps the number of requests, spend limits track actual dollar spend based on token usage and model pricing. Developers can scope limits by model, provider, or custom metadata dimensions, such as giving each user a $200/day budget, capping total gateway spend at $10,000/day, or limiting a specific model to $50/day per user. Each rule uses a configurable time window with fixed or sliding enforcement. Spend limits work with both Unified Billing and BYOK requests for models with known pricing. This feature helps developers prevent unexpected cost overruns when using AI models through the gateway.
Prevents runaway AI costs by enforcing budgets based on actual token usage.