Cencori ships a first-party adapter for @tanstack/ai. Use it when you want TanStack's framework-agnostic AI primitives with Cencori's routing, security, and observability.

Installation

Codetext

npm install cencori @tanstack/ai

Default Adapter

Codetext

import { cencori } from 'cencori/tanstack';
 
const adapter = cencori('gpt-4o');

The default adapter reads CENCORI_API_KEY from the environment.

Custom Provider

Codetext

import { createCencori } from 'cencori/tanstack';
 
const provider = createCencori({
  apiKey: process.env.CENCORI_API_KEY!,
});
 
const adapter = provider('gpt-4o');

Basic Streaming

Codetext

import { chat } from '@tanstack/ai';
import { cencori } from 'cencori/tanstack';
 
for await (const chunk of chat({
  adapter: cencori('gpt-4o'),
  messages: [{ role: 'user', content: 'Write a haiku about coding.' }],
})) {
  if (chunk.type === 'content') {
    process.stdout.write(chunk.delta);
  }
}

Switching Models

Codetext

await chat({
  adapter: cencori('gpt-4o'),
  messages,
});
 
await chat({
  adapter: cencori('claude-sonnet-4.5'),
  messages,
});
 
await chat({
  adapter: cencori('gemini-2.5-flash'),
  messages,
});

Structured Output And Tools

The adapter forwards tool definitions and structured output requests through Cencori, so TanStack AI workflows still benefit from gateway policies and logging.