Comparison

Routeplane vs Cloudflare AI Gateway

Cloudflare gives you a managed AI proxy on a global edge in minutes. Routeplane is a data plane you own, run in-region, and can audit.

Cloudflare AI Gateway is the lowest-friction on-ramp in the category: swap a URL, get caching, analytics, rate limiting, and a generous free tier on a 300+ city edge network, with no infrastructure at all. The 2025 refresh added real substance, dynamic routing, unified billing, BYOK key storage, moderation guardrails, and DLP scanning. If your stack is already Workers-first, it’s a natural default.

The structural trade-off is that it is a closed, managed service on Cloudflare’s global network. You can’t self-host it, read its source, or pin an individual request’s data path to a jurisdiction, the global edge is the product. Routeplane is the opposite bet: an Apache-2.0-core Rust data plane you run in your own cloud and region, with per-request regulated-data classification, region-locked routing, deterministic PII redaction, an MCP gateway for agent tool calls, and a hash-chained audit trail you can hand to a regulator.

Feature-by-feature

CapabilityrouteplaneCloudflare AI Gateway
OpenAI-compatible API unified endpoint
Self-hostable / runs in your cloud managed only
Open-source data plane Apache-2.0 CE closed source
Per-request regulated-data classification
Region-locked routing enforced per request global edge by design
Guardrails / content moderation deterministic PII + checks moderation + DLP
MCP / agent tool-call governance separate products
Global edge cache (300+ PoPs) in-deployment cache
Zero-infrastructure free tier free self-host CE generous hosted tier
Multi-currency FinOps & chargeback
Tamper-evident audit trail hash-chained log push
full partial / tier-gated not available Public-materials snapshot (mid-2026), re-verify before procurement.

Which should you choose?

Choose routeplane when

  • Your data can’t transit a third party’s global network, you need the gateway in your own cloud and region.
  • A regulator cares about per-request residency, not the vendor’s regional-services program.
  • Agents call tools through MCP and you need default-deny governance in the same data plane.
  • You want an open-source core you can read, self-host, and audit.

Choose Cloudflare AI Gateway when

  • You’re already on Cloudflare Workers and want AI caching, analytics, and billing with zero infrastructure.
  • Globally distributed cache-hit latency matters more than where each request’s data travels.
  • A free hosted tier for prototyping is the priority.

Migrating to routeplane

Both products are adopted the same way, a base-URL change on an OpenAI-compatible client, so switching is symmetric. Point your SDKs at https://api.routeplane.ai/v1 (or your self-hosted CE deployment) and keep your code; residency and guardrail decisions come back as x-routeplane-* headers.

Frequently asked questions

Can Cloudflare AI Gateway enforce data residency?

Its architecture routes traffic through Cloudflare’s global network, that’s the point of the product. Cloudflare offers regional-services controls at the platform level, but the AI Gateway does not classify a request’s data and hard-lock that request to an in-jurisdiction model provider. Routeplane does exactly that, per request.

Is Cloudflare’s cache faster than Routeplane’s?

For globally distributed cache hits, honestly, yes: 300+ edge locations are hard to beat. Routeplane caches within your deployment, which keeps cached responses inside your compliance boundary, a different trade-off, chosen deliberately.

Do I have to choose between them?

Not necessarily. Some teams keep Cloudflare in front for CDN/WAF concerns and run Routeplane as the governed AI data plane behind it. The comparison here is specifically about the AI gateway layer.

See the residency header come back true.

Point your existing OpenAI-compatible client at routeplane and route your first sovereign request this week.