Models & plans

Choosing a model

Put a public model ID in the request body's model field, e.g.:

{ "model": "gpt-4o-mini", "messages": [...] }

• OpenAI-compatible and Claude APIs: model goes in the body.
• Gemini API: the model goes in the path, e.g. /v1beta/models/gemini-1.5-pro:generateContent.

Common example model IDs: gpt-4o-mini, text-embedding-3-small, claude-3-5-sonnet, gemini-1.5-pro. The actual available list depends on your account / key — query it via the endpoint below or the console.

List available models

curl https://gateway.mindproxy.ai/v1/models \
  -H "Authorization: Bearer $TT_API_KEY"

Returns an OpenAI-style list containing only the models your key is allowed to call:

{
  "object": "list",
  "data": [
    { "id": "gpt-4o-mini", "object": "model", "owned_by": "tt" },
    { "id": "claude-3-5-sonnet", "object": "model", "owned_by": "tt" }
  ]
}

The console "Model Prices" page shows each model's input / output / cache prices, capabilities and multiplier.

Models & plans

The platform organizes available models and prices into plans. Two things to understand:

1. Each API key is bound to one plan (assigned by the platform). Which plan you use is automatic — it follows the key.
2. The plan decides which models you can use and the price multiplier: price = base price × plan multiplier (1.00x means base price). The same model may be priced differently under different plans.

Current version

The plan is fixed per key and selected automatically. The current version does not support switching plans within a single request. To switch plans, use a key bound to that plan, or contact support to change the key binding.

The console "Model Prices" page has a plan switcher that previews models and prices under different plans, but actual billing still uses the plan your key is bound to.

Capability matrix

Different models support different capabilities:

Capability	Description
Thinking	Reasoning / thinking mode
Prompt cache	Prompt caching; hits billed at the cache price
Response cache	Platform response cache
Image	Multimodal image input / image output

Which a model supports is shown by capability tags on the console "Model Prices" page.