Models

Understanding model availability and naming conventions.

Model Naming Convention

Because InferHood aggregates models from multiple providers, we use a specific naming convention to ensure there are no collisions. Models are referenced using the format: provider/model-name.

For example:

openai/gpt-4o
anthropic/claude-3-opus
google/gemini-pro

Listing Models

You can dynamically retrieve the list of all currently available models, along with their pricing and context windows, using the Models endpoint.

bash

curl https://api.inferhood.xyz/v1/models \
  -H "Authorization: Bearer $INFERHOOD_API_KEY"

json

{
  "object": "list",
  "data": [
    {
      "id": "anthropic/claude-sonnet-4",
      "object": "model",
      "created": 1686935002,
      "owned_by": "anthropic",
      "pricing": {
        "prompt": "0.003",
        "completion": "0.015"
      },
      "context_length": 200000
    }
  ]
}

Pricing

Pricing is determined by the underlying provider plus a small InferHood routing fee. You are billed based on the number of input tokens (prompt) and output tokens (completion) used in each request. For detailed, up-to-date pricing, always check the Pricing page or use the /v1/models endpoint.