Documentation
Strategies
Lower cost, route coding-agent turns, compare models, and cache repeated work.
In this section
Cost-efficient routing
Use lower-cost Flex capacity when it meets your latency needs.
Switchyard routing
Route coding-agent turns between capable and efficient model tiers.
Benchmark routing
Route through weighted benchmark aliases with automatic fallback.
Shadow models
Stop overpaying for every request. Find the cheapest model that performs on your real production traffic.
Prompt caching
Reuse provider-cached prompt prefixes to reduce latency and cost.
