Error with @ai-sdk/google using streamText

Help
,
1

I’m encountering an issue when using the @ai-sdk/google integration in the Vercel AI SDK with the gemini-2.0-flash-001 model. I’m trying to use the streamText function along with tools and the experimental_output option.

code:

import { createGoogleGenerativeAI } from "@ai-sdk/google";
import { Output, smoothStream, streamText, tool } from "ai";
import { z } from "zod";

const model = createGoogleGenerativeAI({
  apiKey: EncodedVideoChunk.GEMINI_API_KEY,
});

const Schema = z.object({
  response: z.string(),
});

const getWeatherInformation = tool({
  description: "show the weather in a given city to the user",
  parameters: z.object({ city: z.string() }),
  execute: async ({ city }) => {
    console.log(`Getting weather information for ${city}`);
    return `The weather in ${city} is sunny`;
  },
});

export const tools = {
  getWeatherInformation,
};

const result = streamText({
  model: model("gemini-2.0-flash-001"),
  messages: [
    { role: "system", content: "You are a helpful assistant." },
    { role: "user", content: "Give me weather of gujarat" },
  ],
  tools,
  maxSteps: 2,
  onError: (error) => {
    console.error("Error while streaming:", JSON.stringify(error, null, 2));
  },
  onFinish: (result) => {
    console.info("Output: " + JSON.stringify(result.object, null, 2));
    console.info("Tool calling: " + JSON.stringify(result.toolCalls, null, 2));
  },
  experimental_transform: smoothStream(),
  experimental_output: Output.object({
    schema: Schema,
  }),
});

for await (const delta of result.textStream) {
  console.info("------------------->", delta);
  console.info("\n\n");
}

However, I’m consistently getting the following error:

{
  "error": {
    "code": 400,
    "message": "For controlled generation of only function calls (forced function calling), please set 'tool_config.function_calling_config.mode' field to ANY instead of populating 'response_mime_type' and 'response_schema' fields. For more details, see: https://cloud.google.com/vertex-ai/generative-ai/docs/multimodal/function-calling#tool-config",
    "status": "INVALID_ARGUMENT"
  }
}

Interestingly, when I use the OpenAI integration with a similar setup (streaming text with tools), it works without any issues.

Could you please help me understand what might be causing this error with the Google Gemini model and how I can resolve it? It seems like there might be a specific configuration requirement for Gemini when using tools and experimental_output with streamText that I’m missing.

Any guidance or suggestions would be greatly appreciated.

Thanks!

2

Did you know about these other v0 resources? A human should be here soon to help, but the answer may already be available even faster in one of these other posts.

This guide is a great place to start.

And these recordings can give you a look at v0 features and strategies in action: Topics tagged v0

3

Same issue. Currently google-genai api raise error when responseSchema is given and function calling mode is ANY or AUTO.

4

Perhaps this might help streamText Tool Invocation Failure - #2 by jessyan0913