OhMyGPT

OhMyGPT provides a unified, OpenAI-compatible API that routes requests to hundreds of AI models. You can use the OpenAI SDK, any OpenAI-compatible client, or raw HTTP requests.

Prerequisites

Requirement	Details
Account	Register at next.ohmygpt.com
API Key	Create one at API Keys
Credit	Add balance via the Billing page

Protect your API keys. Never commit them to public repositories. Use environment variables to keep keys out of your codebase.

Base URL

All API requests go through a single base URL:

https://api.ohmygpt.com/v1

For users in different regions, alternative endpoints are available:

Endpoint	Description	Recommended for
`https://api.ohmygpt.com`	Direct US server	Users in the US
`https://apic.ohmygpt.com`	Cloudflare CDN (Pro)	Users outside the US
`https://c-z0-api-01.hash070.com`	China-optimized CDN	Users in mainland China

Send your first request

from openai import OpenAI

client = OpenAI(
    base_url="https://api.ohmygpt.com/v1",
    api_key="<OHMYGPT_API_KEY>",
)

completion = client.chat.completions.create(
    model="gpt-4o",
    messages=[
        {"role": "user", "content": "What is the meaning of life?"}
    ]
)

print(completion.choices[0].message.content)

Response structure

A successful response contains:

{
  "id": "chatcmpl-abc123",
  "object": "chat.completion",
  "created": 1700000000,
  "model": "gpt-4o",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "The meaning of life is..."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 14,
    "completion_tokens": 125,
    "total_tokens": 139
  }
}

Field	Description
`choices[0].message.content`	The model's response text
`usage.total_tokens`	Total tokens consumed (used for billing)
`finish_reason`	Why the model stopped: `stop`, `length`, or `tool_calls`

Common errors

Code	Meaning	Solution
401	Invalid API key	Check that your key is correct and active
402	Insufficient balance	Add credit at Billing
429	Rate limited	Reduce request frequency or upgrade your tier
502	Model unavailable	Try a different model or wait and retry

Next steps

Models

Browse available models and routing options

Streaming

Stream responses for real-time output

Tool Calling

Let models call external functions

Quickstart