Name: Cencori
Rating: 5 (1 reviews)

Streaming provides real-time responses, showing tokens as they're generated rather than waiting for the complete response.

SDK Streaming

Codetext

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

HTTP Streaming

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": "Tell me a story"}],
    "stream": true
  }'

Stream Events

Event	Description
`delta`	Text content chunk
`tool_call`	Tool call request
`usage`	Token usage stats
`done`	Stream complete

Codetext

for await (const chunk of stream) {
  if (chunk.type === 'delta') {
    console.log('Text:', chunk.delta);
  } else if (chunk.type === 'tool_call') {
    console.log('Tool:', chunk.toolCall);
  } else if (chunk.type === 'usage') {
    console.log('Usage:', chunk.usage);
  }
}

Abort Streaming

Cancel a stream mid-generation:

Codetext

const controller = new AbortController();
 
const stream = cencori.ai.chatStream({
  model: 'gpt-4o',
  messages: [{ role: 'user', content: 'Write a very long essay' }],
  signal: controller.signal
});
 
// Cancel after 2 seconds
setTimeout(() => controller.abort(), 2000);

React Integration

Codetext

'use client';
 
function StreamingChat() {
  const [content, setContent] = useState('');
 
  async function handleSubmit() {
    const stream = await fetch('/api/chat', {
      method: 'POST',
      body: JSON.stringify({ message: 'Hello' }),
    });
 
    const reader = stream.body?.getReader();
    const decoder = new TextDecoder();
 
    while (true) {
      const { done, value } = await reader!.read();
      if (done) break;
      setContent(prev => prev + decoder.decode(value));
    }
  }
 
  return <div>{content}</div>;
}