Vigthoria WebSocket API Documentation

🔌 Connection

WebSocket Endpoint

wss://ws.vigthoria.io/v1

Connecting with JavaScript

const ws = new WebSocket('wss://ws.vigthoria.io/v1');

// Authentication
ws.onopen = () => {
    ws.send(JSON.stringify({
        type: 'auth',
        token: 'YOUR_API_KEY'
    }));
};

// Handle messages
ws.onmessage = (event) => {
    const message = JSON.parse(event.data);
    console.log('Received:', message);
};

// Handle errors
ws.onerror = (error) => {
    console.error('WebSocket error:', error);
};

// Handle disconnection
ws.onclose = () => {
    console.log('Disconnected');
};

Connection Status

Connected

Disconnected

📨 Message Types

Authentication

SEND auth

{
    "type": "auth",
    "token": "YOUR_API_KEY"
}

RECEIVE auth_success

{
    "type": "auth_success",
    "user_id": "user_abc123",
    "plan": "pro"
}

Streaming Chat

SEND chat.stream

Start a streaming chat request:

{
    "type": "chat.stream",
    "message": "Explain quantum computing",
    "model": "vigthoria-reasoning",
    "conversation_id": "conv_xyz789"
}

RECEIVE chat.chunk

Receive streaming chunks:

{
    "type": "chat.chunk",
    "content": "Quantum computing ",
    "done": false
}
// ... more chunks ...
{
    "type": "chat.chunk",
    "content": "qubits.",
    "done": true,
    "tokens_used": 450
}

Generation Progress

SEND subscribe.job

Subscribe to job progress updates:

{
    "type": "subscribe.job",
    "job_id": "job_abc123"
}

RECEIVE job.progress

{
    "type": "job.progress",
    "job_id": "job_abc123",
    "progress": 45,
    "status": "processing",
    "eta_seconds": 30
}

RECEIVE job.completed

{
    "type": "job.completed",
    "job_id": "job_abc123",
    "output_url": "https://cdn.vigthoria.io/..."
}

💬 Streaming Chat Example

const ws = new WebSocket('wss://ws.vigthoria.io/v1');

ws.onopen = () => {
    // Authenticate
    ws.send(JSON.stringify({ type: 'auth', token: API_KEY }));
};

ws.onmessage = (event) => {
    const msg = JSON.parse(event.data);
    
    if (msg.type === 'auth_success') {
        // Start streaming chat
        ws.send(JSON.stringify({
            type: 'chat.stream',
            message: 'Write a poem about AI',
            model: 'vigthoria-creative'
        }));
    }
    
    if (msg.type === 'chat.chunk') {
        // Append chunk to output
        process.stdout.write(msg.content);
        
        if (msg.done) {
            console.log('\\n\\nTokens used:', msg.tokens_used);
        }
    }
};

🔄 Heartbeat

The server sends a ping every 30 seconds. Respond with a pong to keep the connection alive:

ws.onmessage = (event) => {
    const msg = JSON.parse(event.data);
    
    if (msg.type === 'ping') {
        ws.send(JSON.stringify({ type: 'pong' }));
    }
};

⚙️ Features

Streaming Responses
Get AI responses as they're generated, token by token, for instant feedback.
Job Progress
Subscribe to generation jobs and receive real-time progress updates.
Live Notifications
Receive instant notifications for account events, credits, and more.
Low Latency
Direct WebSocket connection for minimal latency compared to polling.

❌ Error Handling

RECEIVE error

{
    "type": "error",
    "code": "rate_limited",
    "message": "Too many requests. Please slow down.",
    "retry_after": 60
}

Common error codes: auth_failed, rate_limited, invalid_message, subscription_required

⚡ WebSocket API