API Overview

Korad.AI provides a fully Anthropic-compatible API — if your code works with Anthropic's API, it works with Korad.AI.

Base URL

https://api.korad.ai/v1

Authentication

Include your API key in the x-api-key header:

curl https://api.korad.ai/v1/messages \
  -H "x-api-key: sk-korad-YOUR-KEY" \
  -H "anthropic-version: 2023-06-01" \
  -H "content-type: application/json" \
  -d '{"model": "claude-sonnet-4-20250514", "max_tokens": 1024, "messages": [{"role": "user", "content": "Hello!"}]}'

Or use the Authorization: Bearer header:

curl https://api.korad.ai/v1/messages \
  -H "Authorization: Bearer sk-korad-YOUR-KEY" \
  ...

Available Endpoints

Endpoint	Description
POST /v1/messages	Create a message (chat completions)
POST /v1/messages/batches	Create batched message requests
GET /v1/models	List available models
POST /v1/messages/{message_id}/streams	Stream responses

Supported Models

Korad.AI supports all Anthropic models plus cost-optimized alternatives:

Model	Context	Best For
claude-sonnet-4-20250514	200K	General tasks, coding
claude-sonnet-4-20250514-optim	128K	Cost-optimized general tasks
claude-haiku-4-20250514	128K	Fast responses, simple tasks
claude-opus-4-20250514	200K	Complex reasoning

Request Format

All requests follow Anthropic's format:

{
  "model": "claude-sonnet-4-20250514",
  "max_tokens": 1024,
  "messages": [
    {
      "role": "user",
      "content": "What is quantum computing?"
    }
  ],
  "temperature": 0.7,
  "top_p": 0.9,
  "stream": false
}

Response Format

{
  "id": "msg_xxxxxxxxx",
  "type": "message",
  "role": "assistant",
  "content": [
    {
      "type": "text",
      "text": "Quantum computing is..."
    }
  ],
  "model": "claude-sonnet-4-20250514",
  "stop_reason": "end_turn",
  "usage": {
    "input_tokens": 15,
    "output_tokens": 150,
    "total_tokens": 165
  },
  "korad_metrics": {
    "savings_percent": 45,
    "original_cost": 0.004,
    "your_cost": 0.002
  }
}

Korad.AI Extensions

We include additional metrics in every response:

Field	Description
korad_metrics.savings_percent	Percentage saved vs Anthropic direct
korad_metrics.original_cost	What this would cost via Anthropic
korad_metrics.your_cost	Actual cost via Korad.AI

Rate Limits

Tier	Requests/Minute	Tokens/Minute
Free	60	40,000
Basic ($50)	300	200,000
Pro ($100)	1000	1,000,000
Enterprise	Custom	Custom

Error Codes

Code	Description
401	Invalid API key
429	Rate limit exceeded
500	Internal server error
502	Upstream API error

View all error codes →

Streaming

Enable streaming with "stream": true:

with client.messages.stream(
    model="claude-sonnet-4-20250514",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Tell me a story"}]
) as stream:
    for text in stream.text_stream:
        print(text, end="", flush=True)

Next Steps

• Messages API → — Detailed message endpoint reference
• Models → — Available models and capabilities
• Error Codes → — Complete error reference