Generative Model API

Welcome to the Generative Model API, a powerful suite of RESTful endpoints designed to seamlessly integrate advanced artificial intelligence into your applications through three core services: Large Language Models (LLMs) for complex text generation, reasoning, and conversational agents; Large Multimodal Models (LMMs) for sophisticated visual processing, such as next-generation OCR capable of digitizing messy or unstructured documents; and Embedding models that translate text into dense vector coordinates to power semantic search and matchmaking. Built for high-throughput, stateless execution via standard JSON payloads, this API provides developers with the foundational building blocks needed to quickly deploy intelligent, context-aware features.

Use this Generative Model API to create text embeddings, generate chat completions, process multimodal inputs, and list the available models.

Authentication

Every request must include your API key as a Bearer token:

Authorization: Bearer YOUR_API_KEY

Base URL:

[servers]

Usage examples

Create an embedding

curl --location '[servers]/v1/embeddings' \
  --header 'Content-Type: application/json' \
  --header 'Authorization: Bearer YOUR_API_KEY' \
  --data '{
    "model": "Qwen3-Embedding-8B",
    "input": "Hello, how are you?"
  }'

Create a chat completion

curl --location '[servers]/v1/chat/completions' \
  --header 'Content-Type: application/json' \
  --header 'Authorization: Bearer YOUR_API_KEY' \
  --data '{
    "model": "gemma-4-26B-A4B-it",
    "messages": [
      {
        "role": "user",
        "content": "Hello, how are you?"
      }
    ]
  }'

Create a multimodal chat completion

Use a model that supports multimodal input. The image can be supplied using a public URL or a Base64 data URL.

curl --location '[servers]/v1/chat/completions' \
  --header 'Content-Type: application/json' \
  --header 'Authorization: Bearer YOUR_API_KEY' \
  --data '{
    "model": "MULTIMODAL_MODEL_NAME",
    "messages": [
      {
        "role": "user",
        "content": [
          {
            "type": "text",
            "text": "Act as an accountant. Read this image and return a JSON object containing the store name, total amount, date, and an array of all line items."
          },
          {
            "type": "image_url",
            "image_url": {
              "url": "https://example.com/receipt.jpg"
            }
          }
        ]
      }
    ]
  }'

For a Base64-encoded image, set image_url.url to a data URL such as data:image/jpeg;base64,BASE64_IMAGE_DATA.

List the available models

curl --location '[servers]/v1/models' \
  --header 'Accept: application/json' \
  --header 'Authorization: Bearer YOUR_API_KEY'