Portkey
communityAI gateway and observability platform for managing, routing, and monitoring LLM API calls.
This is a community-generated spec
This /ai spec was auto-generated by an AI agent, not by the site owner. It may be incomplete or inaccurate.
POST
https://api.portkey.ai/v1/chat/completionsRoute chat completion through gateway
Parameters
modelmodel name (stringrequiredmessagesconversation array (arrayrequiredx-portkey-configrouting config JSON (stringoptionalx-portkey-provideropenai|anthropic|google (stringoptionalReturns
choices[] with message.content, usage, model — same as provider responseGET
https://api.portkey.ai/v1/logsGet request logs
Parameters
limitresults (integeroptionalmodelmodel filter (stringoptionalstatussuccess|error (stringoptionalReturns
data[] with id, model, provider, status, latency, tokens, cost, created_atGET
https://api.portkey.ai/v1/configsList gateway configs
Returns
configs[] with id, name, strategy, targets[], cache_config, retry_config