Model Discovery and Catalog

Inference Space provides three APIs for discovering available models and looking up prices:

• Models the current Key can call: GET /v1/models, OpenAI compatible, and the one used most often during integration.
• Public batch model and price lookup: No authentication required, CDN-friendly, and suitable for website and pricing-page displays.
• Organization-scoped real-time pricing: Requires a gateway API Key and returns the effective price catalog for the organization that owns the Key.

Models the current Key can call (/v1/models)

Returns the model IDs that the organization owning this Key can actually call; the OpenAI SDK's client.models.list() works with it directly. The Key needs one of the ai:llm, ai:image, ai:video, or ai:* scopes; otherwise the request returns 403 insufficient_scope.

# China region, accelerated route; China international route is https://global.inf.space, Global region is https://ai.inf.space
curl "https://cn.inf.space/v1/models" \
  -H "Authorization: Bearer $INFERENCE_SPACE_API_KEY"

{
  "object": "list",
  "data": [
    { "id": "gpt-5.6-sol", "object": "model", "created": 0, "owned_by": "..." },
    { "id": "gpt-image-2", "object": "model", "created": 0, "owned_by": "..." }
  ]
}

• Query a single model: GET /v1/models/{model}.
• Gemini tools can use GET /v1beta/models to get a Gemini-shaped list.
• The list only says what is "callable" and contains no prices; for prices, use /v1/pricing/lookup below.

Public batch model and price lookup

Look up model metadata and prices in the requested currency for a batch of model IDs. Authentication is not required; the response includes CORS and cache headers, making the API suitable for direct use by marketing sites and pricing pages.

Global region: use https://ai.inf.space
POST https://cn.inf.space/v1/public/models/lookup?currency=CNY

• The currency query parameter is required and must be a three-letter ISO 4217 currency code such as USD or CNY. Missing or invalid values return 400.
• The body is {"modelIds": [...]} with at most 200 IDs; more than 200 returns 413.
• No Authorization header is required.

Request

# China region, accelerated route; China international route is https://global.inf.space, Global region is https://ai.inf.space
curl "https://cn.inf.space/v1/public/models/lookup?currency=CNY" \
  -H "Content-Type: application/json" \
  -d '{
    "modelIds": ["qwen-turbo", "qwen-max", "non-existent-id"]
  }'

import requests

resp = requests.post(
    # China region, accelerated route; China international route is https://global.inf.space, Global region is https://ai.inf.space
    "https://cn.inf.space/v1/public/models/lookup",
    params={"currency": "CNY"},
    json={"modelIds": ["qwen-turbo", "qwen-max", "non-existent-id"]},
)
print(resp.json())

const resp = await fetch(
  // China region, accelerated route; China international route is https://global.inf.space, Global region is https://ai.inf.space
  "https://cn.inf.space/v1/public/models/lookup?currency=CNY",
  {
    method: "POST",
    headers: { "Content-Type": "application/json" },
    body: JSON.stringify({
      modelIds: ["qwen-turbo", "qwen-max", "non-existent-id"],
    }),
  },
);
console.log(await resp.json());

Response

The response has the shape { models, currency, asOf }. models is a map keyed by the model IDs in the request: a found ID maps to a model entry, and an unknown ID maps to null.

{
  "models": {
    "qwen-turbo": {
      "id": "qwen-turbo",
      "label": "Qwen Turbo",
      "labelEn": "Qwen Turbo",
      "labelZh": "通义千问 Turbo",
      "capabilityId": "llm",
      "contextWindow": 1000000,
      "supportsVision": false,
      "pricing": {
        "currency": "CNY",
        "inputPerMillionTokens": "0.3",
        "outputPerMillionTokens": "0.6",
        "cachedInputPerMillionTokens": null,
        "lastChangedAt": "2026-05-13T09:58:35.973Z"
      }
    },
    "qwen-max": {
      "id": "qwen-max",
      "label": "Qwen Max",
      "labelEn": "Qwen Max",
      "labelZh": "通义千问 Max",
      "capabilityId": "llm",
      "contextWindow": 32768,
      "supportsVision": false,
      "pricing": {
        "currency": "CNY",
        "inputPerMillionTokens": "2.4",
        "outputPerMillionTokens": "9.6",
        "cachedInputPerMillionTokens": null,
        "lastChangedAt": "2026-05-13T09:58:35.973Z"
      }
    },
    "non-existent-id": null
  },
  "currency": "CNY",
  "asOf": "2026-06-16T02:10:24.193Z"
}

Model entry fields:

Organization-scoped real-time pricing

Look up the effective price catalog for the organization that owns the current gateway API Key (system prices merged with organization-specific overrides). Authentication is required, and the scope is the Key's organization.

Global region: use https://ai.inf.space
GET https://cn.inf.space/v1/pricing/lookup?capability=llm&model=claude-sonnet-4-6

• Every route domain serves this endpoint (cn.inf.space / global.inf.space / ai.inf.space) at the same path.
• Authenticate with Authorization: Bearer gk_...; results are scoped to the organization that owns the Key.
• Use the capability and model query parameters as filters (combine them as needed).
• The response has the shape { currency, entries, version }.

# China region, accelerated route; China international route is https://global.inf.space, Global region is https://ai.inf.space
curl "https://cn.inf.space/v1/pricing/lookup?capability=llm&model=claude-sonnet-4-6" \
  -H "Authorization: Bearer $INFERENCE_SPACE_API_KEY"

import os, requests

resp = requests.get(
    # China region, accelerated route; China international route is https://global.inf.space, Global region is https://ai.inf.space
    "https://cn.inf.space/v1/pricing/lookup",
    params={
        "capability": "llm",
        "model": "claude-sonnet-4-6",
    },
    headers={"Authorization": f"Bearer {os.environ['INFERENCE_SPACE_API_KEY']}"},
)
print(resp.json())

// China region, accelerated route; China international route is https://global.inf.space, Global region is https://ai.inf.space
const url = new URL("https://cn.inf.space/v1/pricing/lookup");
url.search = new URLSearchParams({
  capability: "llm",
  model: "claude-sonnet-4-6",
}).toString();

const resp = await fetch(url, {
  headers: { Authorization: `Bearer ${process.env.INFERENCE_SPACE_API_KEY}` },
});
console.log(await resp.json());

entries contains the price entries in the organization's effective catalog that match the filters. currency is the organization's billing currency, and version identifies the catalog version.

Choosing between the three APIs

Related

• Quickstart
• Authentication and API keys
• Chat Completions API, Image generation and editing