Rate Limiting Now Available for Telnyx Edge Compute

Rate limiting is now available as a platform-managed binding for Edge Compute functions. Declare per-key request limits in telnyx.toml and check them from your handler with no external service or KV namespace required.

What's new

• Rate limiter bindings (GA): Declare per-key request limits in telnyx.toml with [[ratelimits]] blocks. Each binding is exposed on the handler's env argument, so checking a limit is a single function call.
• Multiple limiters per function: Configure independent budgets for free and paid tiers, different endpoints, or tenant-specific quotas in one function.
• Per-site fixed windows: 10-second and 60-second windows enforced at each Edge Compute site. Platform-managed counters survive pod restarts without any storage setup.
• Type generation: telnyx-edge types generates rate limiter declarations automatically.

Why it matters

Rate limiting joins the existing Edge Compute bindings for KV, StatefulActor, Object Storage, and SQLDB. The binding runs on the same platform as the function, so there is no network hop to an external service. The function calls the binding with a key, gets back a decision, and handles the response. Keys can be a user ID, tenant ID, or composite key like tenant:endpoint for per-route quotas.

What's coming

• Sliding window support: Fixed windows are the only mode in this release. Sliding windows are on the roadmap for finer-grained control.
• Global enforcement: Per-site counters are independent today. Cross-site coordination is planned for workloads that need one budget worldwide.

Getting started

1. Update to telnyx-edge v0.4.0 or later.
2. Add a [[ratelimits]] block to your telnyx.toml with a name, limit, and period.
3. Call env.YOUR_NAME.limit({ key: "..." }) from your handler to get a decision.
4. Handle the response in your function: return a 429, fall back to cached data, or degrade gracefully.

Learn more in the Rate Limiting docs, try the Quick Start, or explore the API Reference.