API Access

Welcome to the Fugu Ultra API documentation. Our platform provides standard HTTP endpoints for accessing Fugu Ultra model capabilities.

Quickstart

To use our API, configure your client to use the following Base URL:

• Production: https://fuguultra.net/api/v1

We currently support the following standard HTTP endpoints:

• GET /models: List available models.
• POST /chat/completions: Generate conversational responses.
• POST /responses: Responses endpoint.

Supported models:

• fugu-ultra
• fugu-ultra-20260615

Create an API key

Authentication is handled via standard Bearer tokens. You can create your API keys in the dashboard under Settings > API Keys.

Include your API key in the Authorization header of your requests:

Authorization: Bearer <Your_API_Key>

Make your first HTTP request

You can use any HTTP client to interact with the API. Here is an example of listing available models.

cURL:

curl https://fuguultra.net/api/v1/models \
  -H "Authorization: Bearer YOUR_API_KEY"

JavaScript (fetch):

const response = await fetch("https://fuguultra.net/api/v1/models", {
  headers: {
    "Authorization": "Bearer YOUR_API_KEY"
  }
});
const data = await response.json();
console.log(data);

Python (requests):

import requests

response = requests.get(
    "https://fuguultra.net/api/v1/models",
    headers={"Authorization": "Bearer YOUR_API_KEY"}
)
print(response.json())

Use Chat Completions API

Generate conversational responses.

cURL:

curl https://fuguultra.net/api/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "fugu-ultra",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

JavaScript (fetch):

const response = await fetch("https://fuguultra.net/api/v1/chat/completions", {
  method: "POST",
  headers: {
    "Content-Type": "application/json",
    "Authorization": "Bearer YOUR_API_KEY"
  },
  body: JSON.stringify({
    model: "fugu-ultra",
    messages: [{ role: "user", content: "Hello!" }]
  })
});
const data = await response.json();
console.log(data.choices[0].message);

Python (requests):

import requests

response = requests.post(
    "https://fuguultra.net/api/v1/chat/completions",
    headers={
        "Content-Type": "application/json",
        "Authorization": "Bearer YOUR_API_KEY"
    },
    json={
        "model": "fugu-ultra",
        "messages": [{"role": "user", "content": "Hello!"}]
    }
)
print(response.json()["choices"][0]["message"])

Use Responses endpoint

cURL:

curl https://fuguultra.net/api/v1/responses \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "fugu-ultra",
    "input": "Hello!"
  }'

Review usage logs

Usage is billed using credits. You can view your consumption in the dashboard under Settings > API Usage and purchase credits in Settings > Billing.

How credits are calculated

Credits are usage units for API calls. Actual usage depends on token usage, cached context, orchestration work, and context length.

Credit Usage Standard

For standard context:

• Input tokens: 6,000 credits / 1M tokens
• Output tokens: 36,000 credits / 1M tokens
• Cached input tokens: 600 credits / 1M tokens

For extended context above 272K:

• Input tokens: 12,000 credits / 1M tokens
• Output tokens: 54,000 credits / 1M tokens
• Cached input tokens: 1,200 credits / 1M tokens

Fugu Ultra may also use orchestration tokens to plan, route, verify, and synthesize responses. These tokens are counted using the same standard as regular input, output, and cached input tokens.

Your final credit usage is rounded up to the nearest whole credit. You can review request-level usage in API Usage.

Common Errors

If a request fails, you may encounter one of these common error codes:

• invalid_api_key: The provided API key does not exist or has been revoked.
• invalid_method: The HTTP method is not supported for this endpoint.
• invalid_request_error: The request body is invalid or missing required fields.
• insufficient_quota: Your account does not have enough credits to process the request.
• model_not_found: The requested model is not supported.
• proxy_disabled: The service is temporarily unavailable. Please try again later or contact support.
• api_error: The request could not be processed. Please try again later.