X Enterprises

fastify-x-ai

Vercel AI SDK plugin for unified access to AI providers — text, streaming, embeddings, and structured output.

fastify-x-ai

Vercel AI SDK plugin providing unified AI provider access with text generation, streaming, embeddings, and structured output.

Installation

npm install @xenterprises/fastify-x-ai

Quick Start

await fastify.register(xAI, {
  provider: 'openai',
  apiKey: process.env.OPENAI_API_KEY,
  defaultModel: 'gpt-4o-mini',
})

API

Text Generation

const text = await fastify.xai.generate({
  prompt: 'Summarize this document: ...',
  model: 'gpt-4o',             // optional override
  maxTokens: 1000,
})

Streaming

fastify.post('/chat', async (request, reply) => {
  const stream = await fastify.xai.stream({
    messages: request.body.messages,
  })
  return reply.send(stream)
})

Structured Output

const result = await fastify.xai.generateObject({
  prompt: 'Extract the address from: ...',
  schema: z.object({
    street: z.string(),
    city: z.string(),
    zip: z.string(),
  }),
})

Embeddings

const embedding = await fastify.xai.embed('Hello world')
// Returns: number[]

AI Context

package: "@xenterprises/fastify-x-ai"
type: fastify-plugin
decorates: fastify.xai
providers: [openai, anthropic, google, cohere]
use-when: AI text generation, document summarization, classification, embeddings
methods:
  generate: prompt → text
  stream: messages → SSE stream (for chat)
  generateObject: prompt + zod schema → typed object
  embed: text → number[]
env-required: provider API key (OPENAI_API_KEY, ANTHROPIC_API_KEY, etc.)
Copyright © 2026