Responses API errors

POST /v1/responses returns errors in OpenAI’s Responses error envelope. Errors that originate at Kindo’s edge use the table below; errors that originate upstream are forwarded verbatim with the upstream body and content-type.

Envelope

{
  "error": {
    "message": "Model 'unknown-model' not found.",
    "type": "invalid_request_error",
    "code": "model_not_found"
  }
}

Status / type / code reference

Status	Type	Common codes	Cause
`400`	`invalid_request_error`	`invalid_body`, `model_not_found`, `unsupported_parameter`	Body fails schema validation, references a missing model, or asks for a feature not yet implemented (e.g. `previous_response_id`).
`401`	—	—	Missing, malformed, or revoked API key. The body is a plain string envelope: `{"error": "Unauthorized"}`.
`403`	`permission_error`	`model_access_denied`	The key authenticated, but the user group does not grant access to the model.
`404`	`not_found`	—	E.g. `GET /v1/responses/{id}` for a response that does not exist or has been pruned.
`429`	`rate_limit_error`	—	Org or user rate limit exceeded. Honors the `Retry-After` header when present.
`500`	`server_error`	`internal_error`	Unexpected Kindo-side failure.
`502`	`server_error`	`upstream_empty_body`	Upstream LiteLLM / provider returned an empty response.

`401` exception

The 401 envelope is the plain-string form {"error": "Unauthorized"} rather than the typed {"error": {"message": "...", "type": "...", "code": "..."}} shape. This matches how Kindo’s API-key middleware reports auth failures across the Chat Completions and Responses APIs.

Mid-stream errors

If stream: true is set and an error occurs after the SSE has started, Kindo emits an error event followed by response.failed:

event: error
data: {"error":{"message":"upstream stream interrupted","type":"server_error","code":"stream_error"}}

event: response.failed
data: {"id":"resp_abc123","status":"failed","object":"response"}

The outer HTTP status stays 200 because headers were already flushed when streaming began. SDK consumers should treat any mid-stream error event as terminal.

See Streaming for the full SSE protocol.

Provider passthrough

When an upstream provider (Anthropic, OpenAI, etc.) returns a 4xx or 5xx with its own body, Kindo forwards that body verbatim, preserving the upstream Content-Type. Your client sees whatever shape the upstream produced — Kindo does not rewrap. The exception is upstream_empty_body, which Kindo emits when the upstream returned an HTTP error with no body so consumers always have a parseable JSON envelope to log.

If you want a single consistent error shape across providers, parse on the outer HTTP status first; only fall back to body inspection when you need provider-specific detail.