Name: Cencori
Rating: 5 (1 reviews)

TanStack AI provides framework-agnostic primitives for building AI applications. Cencori integrates as a backend adapter.

Installation

Codetext

npm install @tanstack/ai cencori

Setup

Create the Cencori adapter:

Codetext

import { createCencoriAdapter } from 'cencori/tanstack';
 
const adapter = createCencoriAdapter({
  apiKey: process.env.CENCORI_API_KEY,
});

Basic Usage

Codetext

import { generateText } from '@tanstack/ai';
 
const result = await generateText({
  adapter,
  model: 'gpt-4o',
  prompt: 'Write a haiku about coding',
});
 
console.log(result.text);

Streaming

Codetext

import { streamText } from '@tanstack/ai';
 
const stream = await streamText({
  adapter,
  model: 'claude-opus-4',
  prompt: 'Explain machine learning',
});
 
for await (const chunk of stream) {
  process.stdout.write(chunk);
}

React Integration

Codetext

import { useChat } from '@tanstack/ai/react';
 
function ChatComponent() {
  const { messages, input, setInput, send } = useChat({
    adapter,
    model: 'gpt-4o',
  });
 
  return (
    <div>
      {messages.map((m, i) => (
        <div key={i}>{m.role}: {m.content}</div>
      ))}
      <input 
        value={input} 
        onChange={(e) => setInput(e.target.value)} 
      />
      <button onClick={() => send()}>Send</button>
    </div>
  );
}

Multi-Provider Support

Switch between any supported provider:

Codetext

// Use different models with the same adapter
await generateText({ adapter, model: 'gpt-4o', prompt: '...' });
await generateText({ adapter, model: 'claude-opus-4', prompt: '...' });
await generateText({ adapter, model: 'gemini-2.5-pro', prompt: '...' });