Models

GET /v1/models returns the catalog in the OpenAI API format, plus each model’s type and its prices in credits. The list follows the Moon AI catalog: read it instead of hard-coding IDs in your code.

GET https://api.realmoon.ai/v1/models
GET https://api.realmoon.ai/v1/models?type=image

• Authentication: an sk-moon-… key with the inference scope (see Authentication).
• Response: a list object whose data holds the models, chat models first, sorted by ID, then media models.
• Billing: none.

Parameter#Link to the Parameter section

Example#Link to the Example section

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

curl "https://api.realmoon.ai/v1/models?type=image" \
  -H "Authorization: Bearer $MOONAI_API_KEY"

import os
from openai import OpenAI

client = OpenAI(base_url="https://api.realmoon.ai/v1", api_key=os.environ["MOONAI_API_KEY"])

for model in client.models.list():
    info = model.model_dump()
    print(info["id"], info.get("type"), info.get("pricing"))

images = client.models.list(extra_query={"type": "image"})
print([m.id for m in images])

import OpenAI from 'openai';

const client = new OpenAI({ baseURL: 'https://api.realmoon.ai/v1', apiKey: process.env.MOONAI_API_KEY });

const all = await client.models.list();
for (const model of all.data) console.log(model.id, model.type, model.pricing);

const images = await client.models.list({ query: { type: 'image' } });
console.log(images.data.map((m) => m.id));

Response#Link to the Response section

{
  "object": "list",
  "data": [
    {
      "id": "anthropic/claude-haiku-4.5",
      "object": "model",
      "created": 1777593600,
      "name": "Claude Haiku 4.5",
      "owned_by": "moon-ai",
      "pricing": {
        "prompt_credits_per_1m": 7000,
        "completion_credits_per_1m": 35000,
        "unit": "credits",
        "per": "1M_tokens"
      },
      "context_length": 200000,
      "type": "chat",
      "endpoints": ["/v1/chat/completions"],
      "types": ["chat"]
    },
    {
      "id": "google/gemini-3.1-flash-image",
      "object": "model",
      "created": 1793577600,
      "name": "Nano Banana 2 (Gemini 3.1 Flash Image)",
      "owned_by": "moon-ai",
      "type": "image",
      "endpoints": ["/v1/images/generations", "/v1/images/edits"],
      "hosting": "provider",
      "pricing": {
        "unit": "image",
        "credits_per_unit": 469,
        "max_credits_per_unit": 1155,
        "billing": "real_cost",
        "variants": [
          { "name": "1K", "credits_per_unit": 469, "max_credits_per_unit": 525 },
          { "name": "2K", "credits_per_unit": 707, "max_credits_per_unit": 770 },
          { "name": "4K", "credits_per_unit": 1057, "max_credits_per_unit": 1155 }
        ]
      },
      "aspect_ratios": ["1:1", "16:9", "9:16", "4:3", "3:4", "3:2", "2:3", "21:9"],
      "image_sizes": ["1K", "2K", "4K"],
      "max_input_images": 8,
      "output_format": "png",
      "types": ["image"]
    }
  ]
}

The list object#Link to the The list object section

Fields common to all models#Link to the Fields common to all models section

Chat models (chat)#Link to the Chat models (chat) section

Models hosted by Moon AI have the same input and output price. A model that reads images, audio or files charges for them at its own rates; the exact cost of each request is available from /v1/generation.

Media models#Link to the Media models section

Every media model has:

Fields specific to each type:

Each type is covered in detail in Images, Videos, Audio, Embeddings and OCR.

Which models are served#Link to the Which models are served section

• The models that are active and open to everyone in Moon AI: the major models on the market (OpenAI, Anthropic, Google, Mistral, Meta, DeepSeek, Qwen…) and Moon AI’s in-house models, such as Moon 6, when they are public.
• Never a model with restricted access, an agent, or a workspace’s private model.
• Never a model whose price is not known, or a model that runs paid web searches on every request: their cost cannot be capped in advance.
• In a managed organization, the models its rules refuse are removed from the list.

An ID that is not in the list returns 404 model_not_found. Retrieving a single model (GET /v1/models/{id}) is not supported (404 unknown_url): filter the list instead.

Prices#Link to the Prices section

Prices are in credits. They follow the model’s price at its provider and can change with it; read the list again rather than hard-coding them. The “Models” tab of the portal shows the same prices in euros, with search and sorting by price or by context size (open). How credits convert to euros is explained in Costs and wallet.

Errors#Link to the Errors section

Key and rate errors, common to the whole API, are in Errors.

See also#Link to the See also section

• Chat completions: call a chat model.
• Costs and wallet: how prices turn into an invoice.