Book a Demo

Book a Demo

Book a Demo

Ask Synquery Connector

API v0

← All API docs

Integration guides

Errors and retries

Every error has a stable code. Branch on the code, and retry only what is safe to retry.

Errors return a machine-readable code, a human-readable message, and the field at fault where there is one. Branch on code; messages may change.

When to retry

429 rate_limited: wait for the number of seconds in the Retry-After header, then retry.
500 server_error: safe to retry with the same Idempotency-Key.
Other 4xx errors: fix the request before sending it again. Retrying unchanged returns the same error.

Requests experts can't answer

These are not HTTP errors. The request is accepted, then completes with status: failed and a reason such as deadline_missed or no_matching_experts. Failed requests are not charged.

Error codes

invalid_request400#

Missing, mistyped or unknown field.

unauthorized401#

Missing, invalid or revoked API key.

insufficient_credits402#

Not enough credits. Top up, or switch on auto top-up.

limit_exceeded403#

Over a spend or volume limit you set.

forbidden_scope403#

The key's scope doesn't allow this call.

Unknown request_id, or not visible to this key.

idempotency_conflict409#

Same Idempotency-Key with a different body.

follow_up_used409#

The request's follow-up round was already used.

context_blocked422#

Screening found personal, patient or customer-identifying data.

rate_limited429#

Too many requests. Retry after Retry-After seconds.

server_error500#

Safe to retry with the same Idempotency-Key.

Next

Idempotency →

Error shapejson

{
  "error": {
    "code": "context_blocked",
    "message": "context.notes contains a patient name",
    "field": "context.notes"
  }
}

Rate limited · 429bash

HTTP/1.1 429 Too Many Requests
Retry-After: 2
Content-Type: application/json

{"error": {"code": "rate_limited", "message": "Too many requests"}}