API Reference

Model List

List models available to your API key and select a model in requests.

The model APIs let you inspect which models are available to the current API key. Availability can vary by account, group, channel, feature flag and billing configuration, so this page does not maintain a static pricing table.

For live model availability, pricing, tags and capabilities, use the Model Library or the model details in the console.

Endpoints

MethodPathDescription
GET/v1/modelsList models available to the current API key
GET/v1/models/{model}Retrieve OpenAI-compatible metadata for a model, when available

List Models

curl https://api.tensoraxis.ai/v1/models \
  -H "Authorization: Bearer $TENSORAXIS_API_KEY"

Example response:

{
  "success": true,
  "object": "list",
  "data": [
    {
      "id": "gpt-4o",
      "object": "model",
      "created": 1626777600,
      "owned_by": "openai",
      "supported_endpoint_types": ["openai"]
    }
  ]
}

Common supported_endpoint_types values:

ValueMeaning
openaiChat Completions, Completions, Moderations and other OpenAI-compatible text APIs
openai-responseResponses API
embeddingsEmbeddings API
image-generationImage generation or editing
openai-videoOpenAI/Sora-style video task API
jina-rerankRerank API

Retrieve a Model

curl https://api.tensoraxis.ai/v1/models/gpt-4o \
  -H "Authorization: Bearer $TENSORAXIS_API_KEY"

Example response:

{
  "id": "gpt-4o",
  "object": "model",
  "created": 1626777600,
  "owned_by": "openai"
}

If a tenant-specific or custom model appears in /v1/models but has no standalone OpenAI metadata, use the list endpoint and Model Library as the source of truth.

Selecting a Model

Specify the model with the model field in your request. Different models support different endpoint types, so check the Model Library or /v1/models before submitting production traffic.

response = client.chat.completions.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": "Hello!"}],
)

Async APIs such as video generation also use the model field, but their request bodies are different from Chat Completions. See Video Generation.