The AI Gateway acts as a transparent proxy between your application and AI providers. Instead of integrating with OpenAI, Anthropic, and Google separately, you integrate once with Cencori and get:

Multi-Provider Routing: Switch between providers with a single parameter
Automatic Security: PII detection, prompt injection protection, content filtering
Complete Observability: Every request logged with full prompts, responses, and costs
Failover & Reliability: Automatic retries and provider fallback
Cost Tracking: Real-time usage and spend per project

Available Endpoints

Endpoint	Description	Providers
`/api/ai/chat`	Chat completions (streaming)	OpenAI, Anthropic, Google, xAI, Mistral, DeepSeek, Meta
`/api/ai/embeddings`	Vector embeddings	OpenAI, Google, Cohere
`/api/ai/images/generate`	Image generation	OpenAI, Google
`/api/ai/audio/transcriptions`	Speech-to-text	OpenAI (Whisper)
`/api/ai/audio/speech`	Text-to-speech	OpenAI (TTS)
`/api/ai/moderation`	Content moderation	OpenAI

Chat Completions

The primary endpoint for conversational AI. Supports streaming, tool calling, and structured output.

SDK Usage

Codetext

const response = await cencori.ai.chat({
  model: 'gpt-4o',
  messages: [
    { role: 'system', content: 'You are a helpful assistant.' },
    { role: 'user', content: 'Hello!' }
  ],
  temperature: 0.7,
  maxTokens: 1000
});
 
console.log(response.content);
console.log(response.usage); // { prompt_tokens, completion_tokens, total_tokens }

Streaming

Codetext

const stream = cencori.ai.chatStream({
  model: 'claude-opus-4',
  messages: [{ role: 'user', content: 'Tell me a story' }]
});
 
for await (const chunk of stream) {
  process.stdout.write(chunk.delta);
}

Tool Calling

Codetext

const response = await cencori.ai.chat({
  model: 'gpt-4o',
  messages: [{ role: 'user', content: 'What is the weather in Tokyo?' }],
  tools: [{
    type: 'function',
    function: {
      name: 'get_weather',
      description: 'Get weather for a location',
      parameters: {
        type: 'object',
        properties: { location: { type: 'string' } },
        required: ['location']
      }
    }
  }]
});
 
if (response.toolCalls) {
  console.log(response.toolCalls);
}

Direct API Usage

If you prefer making direct HTTP requests:

Codetext

curl -X POST https://cencori.com/api/ai/chat \
  -H "CENCORI_API_KEY: csk_..." \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4o",
    "messages": [{"role": "user", "content": "Hello!"}],
    "stream": false
  }'

Authentication

Every request requires an API key:

Preferred: CENCORI_API_KEY: csk_...
Also accepted: Authorization: Bearer csk_...

API keys are scoped to projects. Create and manage them in the dashboard.