Quick Start

1.1 Authentication (API Keys)

Prakasa uses Bearer Token authentication. All requests must include your API Key in the Authorization header.

1. Generate Key

1. Log in.

2. Navigate to Settings -> API Keys.

3. Click Create new secret key.

   • Optional: Set expiration (e.g., 30d) or budget limit (e.g., $50).

4. Copy the key (sk-prakasa-...). It is only shown once.

2. Permissions

Keys are scoped to your workspace.

• Production: Use separate keys for dev/prod environments.

• Rotation: Supports seamless key rotation without downtime.

---

1.2 Configuration (Base URL)

To integrate Prakasa, simply override the baseURL in your OpenAI-compatible SDK client.

Endpoint

• Production (Global Edge): https://api.prakasa.me/v1

  • Route: Automatically routes to the nearest low-latency node (US/JP/SG).

  • Protocol: Supports HTTP/2 & HTTP/3 (QUIC).

Parameters

Parameter	Value	Description
baseURL	https://api.prakasa.me/v1	Required. Points to Prakasa Gateway.
apiKey	sk-prakasa-...	Your secret key.
model	claude-3-5-sonnet-20240620	Use standard model IDs. No mapping required.

---

1.3 Implementation

cURL

Test your connection instantly:

Bash

curl https://api.prakasa.me/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $PRAKASA_API_KEY" \
  -d '{
    "model": "gpt-4o",
    "messages": [{"role": "user", "content": "Ping"}]
  }'

Python (OpenAI SDK)

Works natively with openai library (v1.0+):

Python

from openai import OpenAI

client = OpenAI(
    api_key="sk-prakasa-...",
    base_url="https://api.prakasa.me/v1"  # <--- Critical Step
)

response = client.chat.completions.create(
    model="claude-3-5-sonnet-20240620",
    messages=[{"role": "user", "content": "Explain Quantum Entanglement"}],
    stream=True  # Supports SSE streaming
)

for chunk in response:
    print(chunk.choices[0].delta.content or "", end="")

Node.js / TypeScript

TypeScript

import OpenAI from "openai";

const openai = new OpenAI({
  apiKey: process.env.PRAKASA_API_KEY,
  baseURL: "https://api.prakasa.me/v1"
});

async function main() {
  const stream = await openai.chat.completions.create({
    model: "gpt-4o",
    messages: [{ role: "user", content: "Hello world" }],
    stream: true,
  });
  
  for await (const chunk of stream) {
    process.stdout.write(chunk.choices[0]?.delta?.content || "");
  }
}

main();