Zylo API Interactive Explorer

Chat & Examples: Core Parameters

Our API acts as a seamless drop-in replacement for OpenAI using our powerful, pre-determined opus-4.6 model. You can use the exact same payloads, parameters, and official SDKs. Note: The model cannot be changed; any other model name provided in the request will be automatically routed to opus-4.6.

📖 Supported Parameters Reference

model Fixed

Must be included in the request, but the value is ignored. The system is locked to opus-4.6.

messages Required

Array of message objects. Supports system, user, and assistant roles. Supports multimodal array content.

temperature Optional

Float between 0.0 and 2.0. Defaults to 1.0. Lower values are more focused, higher values are more creative.

stream Optional

Boolean. Set to true to receive partial message deltas via Server-Sent Events (SSE).

max_tokens Optional

Integer. The maximum number of tokens to generate in the chat completion.

tools Optional

Array of functions. Used to enable Tool Calling/Function Calling for the model to interact with external APIs.

🚀 Live API Test Dashboard

system_prompt string

user_prompt string

stream boolean

Disabled

temperature float

1.0

max_tokens integer

Extra Features (Expand):

🖼️ Activate Multimodal Vision

🛠️ Activate Tool Calling

💬 Assistant Response Idle

⚙️ System Debug Logs

JSON Payload Preview

Streaming Support (Server-Sent Events)

By enabling the stream parameter, Zylo API delivers responses progressively using Server-Sent Events (SSE). This reduces the perceived latency significantly, allowing the user to read text as soon as it is generated.

The Streaming Mechanism

1. Headers

The HTTP response header has Content-Type: text-event-stream and keeps the connection active until completion.

2. Delta Packets

Every generated fragment is sent back instantly inside a JSON package starting with the prefix data: .

3. Termination Signal

When generation completes, Zylo API sends a final data: [DONE] message and terminates the request.

Standard Stream Line Example

data: {
  "id": "chatcmpl-123",
  "object": "chat.completion.chunk",
  "created": 1677652288,
  "model": "opus-4.6",
  "choices": [
    {
      "index": 0,
      "delta": {
        "content": "Hello"
      },
      "finish_reason": null
    }
  ]
}

data: [DONE]

Multimodal: Vision Capabilities

The Zylo API supports multimodal inputs natively, allowing you to pass both text and images to our fixed model opus-4.6.

JSON: Multimodal Request

{
  "model": "opus-4.6",
  "messages": [
    {
      "role": "user",
      "content": [
        { 
          "type": "text", 
          "text": "What's in this image?" 
        },
        { 
          "type": "image_url", 
          "image_url": { 
            "url": "https://example.com/image.jpg" 
          } 
        }
      ]
    }
  ]
}

Structure Breakdown

🤖 Fixed Model

The model is hardcoded and fixed to opus-4.6. You must include it in the payload, but any changes to this string will be ignored.

📝 Array Content Type

Instead of a simple text string, the content field becomes an array of objects when sending mixed media.

🖼️ Image URL Object

Images are passed using objects with "type": "image_url". You can provide public URLs or Base64 encoded data.

Tool Calling (Function Calling)

Equip the API with the ability to execute external functions by describing your tools in the JSON payload.

Defining a Tool Schema

The tools array requires a specific JSON Schema definition. Let's analyze the get_weather example.

type: "function"

Currently, "function" is the only supported tool type.

function.name

The identifier your backend will use to execute the code.

function.description

Crucial for the AI. Explains exactly what the function does so the model knows when to trigger it.

parameters

A strict JSON Schema object detailing the arguments the model must provide.

JSON: Tool Calling Schema

{
  "model": "opus-4.6",
  "tools": [
    {
      "type": "function",
      "function": {
        "name": "get_weather",
        "description": "Get current weather for a location",
        "parameters": {
          "type": "object",
          "properties": {
            "location": {
              "type": "string",
              "description": "City name"
            },
            "unit": {
              "type": "string",
              "enum": ["celsius", "fahrenheit"]
            }
          },
          "required": ["location"]
        }
      }
    }
  ],
  "messages": [
    {
      "role": "user", 
      "content": "What's the weather like in Tokyo?"
    }
  ]
}

Code Examples

Ready to integrate? Because the Zylo API is fully OpenAI compatible, you just need to change the Base URL in your existing code! No API key is required since our secure proxy handles authentication.

🐍 Python (Official OpenAI SDK)

from openai import OpenAI

client = OpenAI(
    base_url="https://opus-api.matiasvergara.workers.dev/v1",
    api_key="not-needed"
)

response = client.chat.completions.create(
    model="opus-4.6",
    messages=[
        {"role": "system", "content": "You are a friendly assistant."},
        {"role": "user", "content": "Hello, Opus!"}
    ]
)

print(response.choices[0].message.content)

💻 JavaScript / Node.js (Official OpenAI SDK)

import OpenAI from 'openai';

const openai = new OpenAI({
  baseURL: 'https://opus-api.matiasvergara.workers.dev/v1',
  apiKey: 'not-needed'
});

async function main() {
  const completion = await openai.chat.completions.create({
    model: 'opus-4.6',
    messages: [
      { role: 'system', content: 'You are a friendly assistant.' },
      { role: 'user', content: 'Hello, Opus!' }
    ],
  });

  console.log(completion.choices[0].message.content);
}

main();

🖥️ cURL (Terminal)

curl -X POST https://opus-api.matiasvergara.workers.dev/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer not-needed" \
  -d '{
    "model": "opus-4.6",
    "messages": [
      {"role": "system", "content": "You are a friendly assistant."},
      {"role": "user", "content": "Hello, Opus!"}
    ]
  }'

Limits & Policy

Important information regarding the availability and usage limits of the Zylo API.

⚠️

Global Rate Limit

There is a strict global limit of 100 requests per day shared across all users. Please use the API responsibly to ensure everyone has fair access.

🕑

Availability Disclaimer

The uptime of this API highly depends on the administrator's personal usage. There is no guarantee that the service will be available every single day. If the global limit is reached or the administrator requires the resources, the API may become temporarily unavailable.