Routing rules now available on AI Gateway
Vercel's AI Gateway just made model swaps frictionless. No code deploys. One rule. Instant reroute.
Why it matters
Teams can now enforce model governance and handle provider outages at the infrastructure layer instead of shipping code—reducing friction in multi-model deployments and lowering switching costs between LLM providers.
The key facts
13 to knowVercel AI Gateway now supports routing rules in beta
Two rule types: Rewrite (swap models transparently) and Deny (block models with 403)
Rules applied at gateway level, not in application code
Instant rerouting when models go down or retire—no code deployment required
Use cases: model unavailability fallback, cost optimization (route expensive to cheaper models), team governance, model migration
Managed via Vercel CLI
All existing gateway configurations (filters, sorting, provider options) apply to destination models
Vercel AI Gateway routing rules now in beta
Two rule types: Rewrite (model substitution) and Deny (model blocking)
Rules applied at gateway level, managed via Vercel CLI
Enables instant model swaps without code deployment
Use cases: model unavailability failover, retirement migration, cost optimization (expensive to cheaper model routing)
Rules inherit all other gateway configurations (data retention, provider allowlist, filters)
Go to the source
Vercel Blogvercel.com
Publisher excerpt: now supports .Vercel AI Gatewayrouting rules Routing rules are firewall-style rules that control which models your team can use, applied at the gateway level instead of in your application code. When a model goes down or gets retired, you usually have to ship a code change to move off it. With…