Skip to main content
Use /v1 for your default router. For a specific router, send the same body to https://routing.dari.dev/{router_id}/v1/responses. The response follows the OpenAI Responses format and adds dari_routing, which identifies the selected model and reasoning effort.

​
Streaming And Tools

Set stream: true for HTTP streaming. Function, custom, and namespace tools use the standard Responses request shape. WebSockets and previous_response_id are not supported; resend the complete input history on each turn.

​
Change GPT-6 Reasoning Effort

GPT-6 models support OpenAI configuration_update input items. Set a request-level reasoning.effort, keep it unchanged across turns, then place an update before the next user message:
Dari routes according to the latest update and forwards the complete update history to the selected GPT-6 Responses model. Keeping the request-level effort stable preserves the earlier prompt-cache prefix. Because previous_response_id is unsupported, include earlier messages and configuration updates in each subsequent request. See the Responses API Reference for supported fields.