{"generatedAt":"2026-08-09","schemaVersion":1,"source":"https://allaiask.com/errors","license":"https://allaiask.com/terms","observationsExportedAt":"2026-08-09","taxonomy":[{"provider":"anthropic","sourceUrl":"https://docs.anthropic.com/en/api/errors","verifiedAt":"2026-08-09","signatures":[{"code":"max_tokens_required","httpStatus":400,"wireSample":"{\"type\":\"error\",\"error\":{\"type\":\"invalid_request_error\",\"message\":\"max_tokens: Field required\"}}","retryable":false,"retryAfter":"none","note":"Messages API has no server-side default for max_tokens — omit it and the request 400s before the prompt is sent."},{"code":"overloaded","httpStatus":529,"wireSample":"{\"type\":\"error\",\"error\":{\"type\":\"overloaded_error\",\"message\":\"Overloaded\"}}","retryable":true,"retryAfter":"none","note":"Anthropic-wide capacity signal, not specific to your key — back off and retry."},{"code":"rate_limited","httpStatus":429,"wireSample":"{\"type\":\"error\",\"error\":{\"type\":\"rate_limit_error\",\"message\":\"Number of request tokens has exceeded your per-minute rate limit\"}}","retryable":true,"retryAfter":"header","note":"Requests and tokens are limited separately — this message tells you which one tripped."}]},{"provider":"openai","sourceUrl":"https://platform.openai.com/docs/guides/error-codes","verifiedAt":"2026-08-09","signatures":[{"code":"param_locked","httpStatus":400,"wireSample":"{\"error\":{\"message\":\"Unsupported value: 'temperature' does not support 0.4 with this model. Only the default (1) value is supported.\",\"type\":\"invalid_request_error\",\"param\":\"temperature\",\"code\":\"unsupported_value\"}}","retryable":false,"retryAfter":"none","note":"Reasoning-mode models ignore a custom temperature and always sample at 1 — the field is accepted but pinned."},{"code":"rate_limited","httpStatus":429,"wireSample":"{\"error\":{\"message\":\"Rate limit reached for requests. Limit 3, Used 3, Requested 1. Please try again in 20s.\",\"type\":\"requests\",\"param\":null,\"code\":\"rate_limit_exceeded\"}}","retryable":true,"retryAfter":"header","note":"The message names which limit tripped (requests or tokens) and the exact wait, unlike a bare 429."}]}],"pages":[{"slug":"anthropic-max-tokens-required","status":"draft","provider":"anthropic","code":"max_tokens_required"},{"slug":"openai-param-locked","status":"draft","provider":"openai","code":"param_locked"},{"slug":"anthropic-overloaded","status":"draft","provider":"anthropic","code":"overloaded"},{"slug":"openai-rate-limited","status":"draft","provider":"openai","code":"rate_limited"},{"slug":"anthropic-rate-limited","status":"draft","provider":"anthropic","code":"rate_limited"}]}