Automatic Fallbacks

​

Overview

​

Server-Side Fallbacks

const response = await client.chat.completions.create({
  model: 'gpt-4o',
  messages: [{ role: 'user', content: 'Hello!' }],
  routing: {
    fallback: ['claude-3.5-sonnet', 'gemini-2.0-flash'],
  },
});

1. Try gpt-4o
2. If it fails (5xx, rate limit, timeout) → try claude-3.5-sonnet
3. If that fails → try gemini-2.0-flash
4. If all fail → return error

​

Client-Side Fallbacks

import OpenModex from '@openmodex/sdk';

const client = new OpenModex({
  apiKey: 'omx_sk_...',
  fallbackModels: ['claude-3.5-sonnet', 'gemini-2.0-flash'],
});

// If the requested model fails, the SDK automatically retries
// with each fallback model
const response = await client.chat.completions.create({
  model: 'gpt-4o',
  messages: [{ role: 'user', content: 'Hello!' }],
});

client := openmodex.NewClient("omx_sk_...",
    openmodex.WithFallbackModels("claude-3.5-sonnet", "gemini-2.0-flash"),
)

​

When Fallbacks Trigger

• 5xx errors — server errors from the provider
• 429 errors — rate limit exceeded
• Timeouts — request didn’t complete in time
• Network errors — connection failures

• 4xx errors (except 429) — these indicate a problem with your request
• Successful responses — even if the content is unexpected

​

Check If Fallback Was Used

const response = await client.chat.completions.create({...});

console.log(response.openmodex?.model_used);     // actual model that responded
console.log(response.openmodex?.fallback_used);   // true if fallback was used