Fehlercodes

Wenn ein Fehler auftritt, gibt die Venice API einen HTTP-Statuscode und eine beschreibende Nachricht zurück. Einige Fehler enthalten zusätzliche Felder, wie z. B. Erstattungsstatus, vorgeschlagene Wiederholungsmodelle, Validierungsdetails oder Request-IDs.

Fehlerantwortformat

Die meisten Fehler geben ein error-Feld mit einer menschenlesbaren Nachricht zurück:

{
  "error": "Invalid request parameters"
}

Einige Endpoints enthalten zusätzliche Felder, die spezifisch für den Endpoint sind. Beispielsweise können Video-Moderationsfehler enthalten, ob Credits erstattet wurden:

{
  "error": "Your generation was blocked due to the model provider's content policies. Credits have been refunded.",
  "credits_refunded": true
}

Validierungsfehler können details- und issues-Felder mit feldbezogenen Validierungsinformationen enthalten:

{
  "error": "Invalid request parameters",
  "details": {
    "_errors": [],
    "field": {
      "_errors": ["Field is required"]
    }
  },
  "issues": []
}

Einige OpenAI-kompatible Endpoints geben für bestimmte Fälle möglicherweise OpenAI-artige Fehlerobjekte zurück, z. B. bei Fehlern wegen Kontextlänge:

{
  "error": {
    "message": "Your request exceeds the model's maximum context. Please reduce your prompt or completion length.",
    "type": "invalid_request_error",
    "param": "messages",
    "code": "context_length_exceeded"
  }
}

Die Fehlermeldungen in der Spalte “Message” werden von der API in englischer Sprache zurückgegeben und sind hier unverändert wiedergegeben.

Error Code	HTTP-Status	Message	Log Level
`AUTHENTICATION_FAILED`	401	Authentication failed	-
`AUTHENTICATION_FAILED_INACTIVE_KEY`	401	Authentication failed - Pro subscription is inactive. Please upgrade your subscription to continue using the API	-
`X402_INVALID_SIGN_IN`	401	Invalid Sign-in-with-x signature	-
`PRO_ONLY_MODEL`	401	This model is only available to Pro users	-
`INSUFFICIENT_BALANCE`	402	Insufficient USD or Diem balance to complete request. Visit https://venice.ai/settings/api to add credits.	-
`API_KEY_DIEM_SPEND_LIMIT_EXCEEDED`	402	API key DIEM spend limit exceeded. Your account may still have DIEM balance, but this API key has reached its configured DIEM spending limit.	-
`API_KEY_USD_SPEND_LIMIT_EXCEEDED`	402	API key USD spend limit exceeded. Your account may still have USD balance, but this API key has reached its configured USD spending limit.	-
`UNAUTHORIZED`	403	Unauthorized access	-
`API_ACCESS_DISABLED`	403	API access has been disabled for this account. Please contact support@venice.ai	-
`X402_WALLET_MISMATCH`	403	You can only access resources for your own wallet	-
`INVALID_REQUEST`	400	Invalid request parameters	-
`INVALID_MODEL`	400	Invalid model specified	-
`REQUEST_ID_NOT_FOUND`	400	Request ID is invalid.	-
`INVALID_AUDIO_FORMAT`	400	Invalid audio format. Supported formats are WAV and MP3.	-
`INVALID_VIDEO_FORMAT`	400	Invalid video format. Supported formats include MP4, MOV, WebM, MKV, AVI, and others.	-
`CORRUPTED_IMAGE`	400	The image file is corrupted or unreadable	-
`IMAGE_TOO_SMALL`	400	Image dimensions are too small	-
`TOO_MANY_TOKENS`	400	Your request exceeds the model’s maximum context. Please reduce your prompt or completion length.	-
`CHARACTER_NOT_FOUND`	404	No character could be found from the provided character_slug	-
`MODEL_NOT_FOUND`	404	Specified model not found	-
`MEDIA_NOT_FOUND`	404	Media could not be found. Request may be invalid, expired, or deleted.	-
`PAYLOAD_TOO_LARGE`	413	The request payload is too large. Please reduce the size of your request.	-
`INVALID_CONTENT_TYPE`	415	Invalid request content-type	-
`VIDEO_DURATION_TOO_LONG`	422	Video duration exceeds the maximum allowed.	-
`VIDEO_DURATION_TOO_SHORT`	422	Video duration is too short.	-
`IMAGE_TOO_LARGE`	422	Image dimensions exceed the maximum allowed.	-
`CONTENT_POLICY_VIOLATION`	422	Your prompt violates the content policy of Venice.ai or the model provider	-
`ASR_UPSTREAM_VALIDATION_FAILED`	422	The audio could not be processed for transcription. Common causes: zero-length, silent, corrupt, or unsupported-language audio. Please verify the file and retry.	warn
`RATE_LIMIT_EXCEEDED`	429	Rate limit exceeded	-
`MODEL_OVERLOADED`	429	The model is currently overloaded. Please try again later.	-
`INFERENCE_FAILED`	500	Inference processing failed	error
`UPSCALE_FAILED`	500	Image upscaling failed	error
`IMAGE_EDIT_ERROR`	500	Image edit failed	error
`UNKNOWN_ERROR`	500	An unknown error occurred	error
`TEE_ATTESTATION_FAILED`	502	TEE attestation request failed. The Trusted Execution Environment provider may be temporarily unavailable.	error
`TEE_SIGNATURE_FAILED`	502	TEE signature request failed. The Trusted Execution Environment provider may be temporarily unavailable.	error
`ASR_UPSTREAM_FAILED`	502	Audio transcription failed due to a temporary upstream error. Please retry.	warn
`MODEL_OFFLINE`	503	The model is temporarily offline. Please try again later.	-
`MODEL_AT_CAPACITY`	503	The model is at capacity. Please try again later.	-
`REQUEST_TIMEOUT`	504	The request took too long to complete and was timed-out. For long-running non-streaming inference requests, use the streaming API by setting `stream=true`. Streaming requests that time out before the first chunk instead report that the model did not start responding in time, usually under heavy load, and to retry or use a different model.	-

​Fehlerantwortformat

​Fehlercodes

Fehlerantwortformat

Fehlercodes