First-Party AI SDK Provider

Vercel AI SDK with DEVUP AI

Build interactive streaming interfaces and AI-native applications using Vercel's ai library with DEVUP AI. Our first-party devupai/ai provider implements the official LanguageModelV3 specification, giving you instant access to generateText, streamText, and Next.js full-stack React streaming hooks.

Get API Key View Provider Documentation

Under the hood: The devupai/ai provider uses @ai-sdk/openai-compatibleto target DEVUP AI's verified OpenAI-compatible API endpoint.

OpenAI Gateway Specs

Architecture Flow

How devupai/ai Integrates with Vercel AI SDK

The devupai/ai subpath export wraps @ai-sdk/openai-compatible to translate high-level Vercel AI SDK calls into compliant HTTP and SSE requests dispatched directly to https://api.devupai.com/v1.

1. App Layer

Vercel AI SDK

generateText, streamText, useChat

2. Provider

devupai/ai

createDevupAI & Model Factory

3. Gateway

DEVUP AI /v1

https://api.devupai.com/v1

4. Execution

Target Model

DeepSeek-V4-Pro, Qwen, etc.

Installation

Install Packages

Install the official devupai SDK along with Vercel's core ai library and its peer dependency @ai-sdk/openai-compatible:

bash

npm install devupai ai @ai-sdk/openai-compatible

Configuration

Initialize createDevupAI Provider

Import createDevupAI from devupai/ai. The provider automatically falls back to your DEVUP_API_KEY environment variable if not passed explicitly:

typescript

import { createDevupAI } from "devupai/ai";

// Initialize the provider with your API key
export const devupai = createDevupAI({
  apiKey: process.env.DEVUP_API_KEY,
  // baseURL defaults to "https://api.devupai.com/v1"
});

Default Instance: If you have DEVUP_API_KEY set in your environment, you can also import the pre-initialized default instance directly: import { devupai } from "devupai/ai";.

Text Generation

One-Shot Generation with generateText

Use standard Vercel AI SDK generateText by passing your initialized provider instance configured with any supported DEVUP AI model identifier:

typescript

import { generateText } from "ai";
import { createDevupAI } from "devupai/ai";

const devupai = createDevupAI({
  apiKey: process.env.DEVUP_API_KEY!,
});

async function main() {
  const { text } = await generateText({
    model: devupai("deepseek-ai/DeepSeek-V4-Pro"),
    prompt: "Explain neural networks in simple terms.",
  });

  console.log(text);
}

main().catch(console.error);

Real-Time Streaming

Streaming with streamText

Stream responses in Node.js or terminal applications using asynchronous iteration over result.textStream:

typescript

import { streamText } from "ai";
import { createDevupAI } from "devupai/ai";

const devupai = createDevupAI({
  apiKey: process.env.DEVUP_API_KEY!,
});

async function main() {
  const result = streamText({
    model: devupai("deepseek-ai/DeepSeek-V4-Pro"),
    prompt: "Write a short story set in Algiers.",
  });

  for await (const text of result.textStream) {
    process.stdout.write(text);
  }
}

main().catch(console.error);

Full-Stack Next.js

App Router Route Handler Streaming

Create an App Router endpoint that returns a streaming SSE response compatible with Vercel's frontend hooks:

typescript

import { streamText } from "ai";
import { createDevupAI } from "devupai/ai";

const devupai = createDevupAI({
  apiKey: process.env.DEVUP_API_KEY!,
});

export async function POST(request: Request) {
  const { messages } = await request.json();

  const result = streamText({
    model: devupai("deepseek-ai/DeepSeek-V4-Pro"),
    messages,
  });

  // Returns standard SSE UI message stream for useChat hooks
  return result.toUIMessageStreamResponse();
}

Provider Specifications

Verified Capabilities Matrix

LanguageModelV3 (Text & Chat)

Exposed as callable devupai(modelId) or via devupai.languageModel(modelId) / devupai.chatModel(modelId).

Compatible with generateText
Compatible with streamText
Compatible with generateObject & structured schemas

Embeddings & Modalities

Exposed via dedicated model constructor methods on the provider instance.

devupai.embeddingModel(...) for embed / embedMany
devupai.imageModel(...) for image generation
devupai.completionModel(...) for legacy completion prompts

Provider Settings Schema (DevupAIProviderSettings)

Option	Type	Default	Description
apiKey	string	process.env.DEVUP_API_KEY	Your DEVUP AI API key from dashboard
baseURL	string	"https://api.devupai.com/v1"	DEVUP AI OpenAI-compatible endpoint URL
headers	Record<string, string>	undefined	Custom HTTP headers attached to every request

Build AI Applications with Vercel AI SDK Today

Get your API key, install devupai/ai, and integrate foundation models directly into your Next.js and React components.

Get API Key View Provider Documentation