OhMyGPT

Auto Router

Set the model to auto and OhMyGPT picks one for each request, priced as that model.

Set model to auto and OhMyGPT picks a model for each request from what the latest user message asks for: a fast, low-cost model for greetings and questions with a short, well-known answer, a capable general model for everyday writing, explanations and ordinary code, and a frontier model for hard problems such as competition math, non-trivial algorithms, system design or tricky debugging. ohmygpt/auto and OpenRouter's openrouter/auto work the same way.

curl https://api.ohmygpt.com/v1/chat/completions \
  -H "Authorization: Bearer $OHMYGPT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "auto",
    "messages": [{"role": "user", "content": "hi"}]
  }'

The response's model names the model that answered:

{
  "model": "deepseek/deepseek-v4.1-flash-1017",
  "choices": [{"message": {"role": "assistant", "content": "Hi! How can I help you today?"}}]
}

You pay that model's standard price; the router adds no fee. Auto works on Chat Completions, Responses, Messages and Gemini, streaming or not. Endpoints that don't generate chat replies, such as Embeddings, reject it with 400 unsupported_endpoint.

Conversations

Send prompt_cache_key and the conversation stays on one model, so its prompt cache keeps working. Within a conversation the router can move up to a stronger model when a later turn is harder, and never moves back down:

TurnModel
hideepseek/deepseek-v4.1-flash-1017
Design a rate limiter for an API gateway that runs in three regions.openai/gpt-6.1-sol
thanksopenai/gpt-6.1-sol

Without prompt_cache_key, each request is routed on its own.

Options

Add an auto-router entry to plugins, as on OpenRouter:

{
  "model": "auto",
  "plugins": [{
    "id": "auto-router",
    "allowed_models": ["openai/*", "google/*"],
    "excluded_models": ["openai/gpt-5.4-nano"],
    "cost_tier": "low"
  }],
  "messages": [{"role": "user", "content": "explain why the sky is blue"}]
}
FieldEffect
allowed_modelsOnly pick from these models. * matches any part of an ID, as in openai/*.
excluded_modelsNever pick these models, even if allowed_models matches them.
cost_tierlow picks one step cheaper than the request would otherwise get, high one step stronger; medium is the default.

A cost_tier other than these, or lists that leave no model, return 400 invalid_parameter naming the field. The auto-router entry is not sent on to the model.

Was this page helpful?

Still stuck? Live chat · [email protected] · Telegram group