Errors

Errors use the OpenAI error shape, so the official SDKs turn them into their normal exceptions (APIError, RateLimitError, AuthenticationError and so on). Branch on code; message is a human-readable explanation in English.

{
  "error": {
    "message": "This request may cost up to $0.0123 but your balance is $0.0040. Add funds at /pricing.",
    "type": "insufficient_quota",
    "code": "insufficient_balance"
  }
}

Failed requests are not charged. Any response with a 4xx or 5xx status costs nothing.

Gateway error codes

Upstream errors

Errors returned by the upstream provider, for example an invalid parameter for a specific model, a context length overflow, a content-policy refusal or a provider's own 429, are passed through with their original status code and the same { error: {... } } shape.

What to do

Handling errors in code

import OpenAI from 'openai'

const client = new OpenAI({ baseURL: 'https://api14.com/api/v1', apiKey: process.env.API14_API_KEY })

try {
  await client.chat.completions.create({ model: 'openai/gpt-5-mini', messages: [{ role: 'user', content: 'Hi' }] })
} catch (err) {
  if (err instanceof OpenAI.APIError) {
    console.error(err.status, err.code, err.message)
    if (err.code === 'insufficient_balance') {
      // prompt the user to add funds
    }
  } else {
    throw err
  }
}

from openai import OpenAI, APIStatusError

client = OpenAI(base_url="https://api14.com/api/v1", api_key="sk-api14-...")

try:
    client.chat.completions.create(model="openai/gpt-5-mini", messages=[{"role": "user", "content": "Hi"}])
except APIStatusError as e:
    print(e.status_code, e.code, e.message)

Errors that occur after a stream has started arrive inside the stream. See Streaming.