@modelfusion/vercel-ai

ModelFusion streams for the Vercel AI SDK.

Usage

ModelFusionTextStream(result: AsyncIterable<string>, callbacks?: AIStreamCallbacksAndOptions)

You can call ModelFusionTextStream with the result from streamText. This stream is compatible with the Vercel AI SDK and supports the callbacks and stream data features.

To convert Vercel AI SDK messages to ModelFusion messages, use the asChatMessages helper.

Example: Ollama & Next.js

This is an example for a Next.js app router API route. It uses the ModelFusion Ollama integration. For a full example app, check out the Next.js, Vercel AI SDK, Ollama & ModelFusion starter.

// app/api/chat/route.ts
import { asChatMessages, ModelFusionTextStream } from "@modelfusion/vercel-ai";
import { Message, StreamingTextResponse } from "ai";
import { ollama, streamText } from "modelfusion";

export const runtime = "edge";

export async function POST(req: Request) {
  const { messages }: { messages: Message[] } = await req.json();

  // Use ModelFusion to call Ollama:
  const textStream = await streamText(
    ollama.ChatTextGenerator({ model: "llama2:chat" }).withChatPrompt(),
    {
      system:
        "You are an AI chat bot. " +
        "Follow the user's instructions carefully.",

      // map Vercel AI SDK Message to ModelFusion ChatMessage:
      messages: asChatMessages(messages),
    }
  );

  // Return the result using the Vercel AI SDK:
  return new StreamingTextResponse(
    // Convert the text stream to a Vercel AI SDK compatible stream:
    ModelFusionTextStream(
      textStream,
      // optional callbacks:
      {
        onStart() {
          console.log("onStart");
        },
        onToken(token) {
          console.log("onToken", token);
        },
        onCompletion: () => {
          console.log("onCompletion");
        },
        onFinal(completion) {
          console.log("onFinal", completion);
        },
      }
    )
  );
}