Error Handling

Build robust applications with proper error handling. The Mume Gateway uses standard HTTP error codes and the OpenAI SDK error classes.

Standard Errors

Python

import openai

client = openai.OpenAI(
    api_key="your-api-key",
    base_url="https://mume.ai/api/v1",
)

try:
    response = client.chat.completions.create(
        model="anthropic/claude-haiku-4.5",
        messages=[{"role": "user", "content": "Hello!"}],
    )
    print(response.choices[0].message.content)
except openai.APIConnectionError as e:
    print(f"Connection error: {e}")
except openai.RateLimitError as e:
    print(f"Rate limit exceeded: {e}")
except openai.APIStatusError as e:
    print(f"API error: {e.status_code} - {e.message}")

JavaScript

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "your-api-key",
  baseURL: "https://mume.ai/api/v1",
});

try {
  const response = await client.chat.completions.create({
    model: "anthropic/claude-haiku-4.5",
    messages: [{role: "user", content: "Hello!"}],
  });
  console.log(response.choices[0].message.content);
} catch (error) {
  if (error instanceof OpenAI.APIConnectionError) {
    console.error("Connection error:", error.message);
  } else if (error instanceof OpenAI.RateLimitError) {
    console.error("Rate limit exceeded:", error.message);
  } else if (error instanceof OpenAI.APIError) {
    console.error(`API error: ${error.status} - ${error.message}`);
  } else {
    throw error;
  }
}

Error Codes

Status	Error Type	Description
400	Bad Request	Invalid request parameters or malformed JSON
401	Unauthorized	Missing or invalid API key
402	Insufficient Credits	Account has no remaining credits
429	Rate Limit	Too many requests — back off and retry
500	Server Error	Internal server error — retry with backoff
503	Service Unavailable	Provider temporarily unavailable

Streaming Error Handling

When using streaming, errors that occur mid-stream are delivered as SSE events rather than thrown exceptions. This allows partial responses to be preserved.

Python

try:
    stream = client.chat.completions.create(
        model="anthropic/claude-haiku-4.5",
        messages=[{"role": "user", "content": "Hello!"}],
        stream=True,
    )
    for chunk in stream:
        # Check for error events in the stream
        if hasattr(chunk, "error") and chunk.error:
            print(f"Stream error: {chunk.error.get('message')}")
            break
        content = chunk.choices[0].delta.content
        if content:
            print(content, end="", flush=True)
except openai.APIError as e:
    # Connection or pre-stream errors are still raised as exceptions
    print(f"API error: {e}")

JavaScript

try {
  const stream = await client.chat.completions.create({
    model: "anthropic/claude-haiku-4.5",
    messages: [{role: "user", content: "Hello!"}],
    stream: true,
  });

  for await (const chunk of stream) {
    if (chunk.error) {
      console.error("Stream error:", chunk.error.message);
      break;
    }
    const content = chunk.choices[0]?.delta?.content;
    if (content) {
      process.stdout.write(content);
    }
  }
} catch (error) {
  if (error instanceof OpenAI.APIError) {
    console.error(`API error: ${error.status} - ${error.message}`);
  } else {
    throw error;
  }
}

Note: Streaming errors include a reqId field that can be used for debugging and support requests.

Best Practices

Always wrap API calls in try/catch blocks
Implement exponential backoff for 429 and 5xx errors
Log the reqId from streaming errors for debugging
Monitor error rates to detect provider issues early
Consider fallback models — if one provider is down, switch to another

← Models Next: Authentication →