Models
List the model catalog with pricing and capability metadata.
List models
GET /v1/modelsReturns the enabled catalog in the OpenAI list shape ({ object: "list", data: [...] }), so stock SDKs work unchanged. Each entry extends the thin OpenAI model
object with the metadata Maniac clients need to route and price intelligently.
const { data } = await client.models.list();Model object
{
// OpenAI base
"id": "openai/gpt-4o-mini",
"object": "model",
"created": 1718380800,
"owned_by": "openai",
// Maniac extensions
"slug": "gpt-4o-mini",
"display_name": "GPT-4o mini",
"provider": "openai",
"context_window": 128000,
"max_output_tokens": 16384,
"pricing": {
"input": 0.00000015, // USD per token
"output": 0.0000006,
"cache_read": 0.0000000375,
"cache_write": 0.0000006
},
"input_modalities": ["text", "image"],
"output_modalities": ["text"],
"supported_parameters": ["temperature", "top_p", "max_tokens"],
"supported_apis": ["chat", "responses"]
}| Field | Meaning |
|---|---|
provider | The upstream that serves the model. |
context_window | Maximum total tokens (input + output). |
max_output_tokens | Maximum tokens the model can emit in one response. |
pricing | Human-facing USD per token (null distinguishes free from unknown). |
input_modalities / output_modalities | Supported content kinds. |
supported_parameters | Sampling/control params honored for this model. |
supported_apis | Which endpoints accept the model (chat, responses). |
Use the model id (e.g. openai/gpt-4o-mini) as the model field in
chat completions and
responses.
See the live schema in the API Reference.