context window — The maximum amount of text a model can consider at once — its working memory for the current conversation or task.
Why it matters
You can set a throughput or latency floor and have slow providers dropped from routing without added latency, constrain the auto router with wildcards such as anthropic/*, and load composable skills into any model's context windowThe maximum amount of text a model can consider at once — its working memory for the current conversation or task.Full definition → from the SDK.
