File: tool-calling-with-structured-outputs.md | Updated: 11/15/2025
Menu
v5 (Latest)
AI SDK 5.x
Model Context Protocol (MCP) Tools
Client-Side Function Calls Not Invoked
Server Actions in Client Components
useChat/useCompletion stream output contains 0:... instead of text
Tool Invocation Missing Result Error
Streaming Not Working When Deployed
Streaming Not Working When Proxied
Getting Timeouts When Deploying on Vercel
useChat Failed to Parse Stream
Server Action Plain Objects Error
Custom headers, body, and credentials not working with useChat
TypeScript performance issues with Zod and AI SDK 5
Repeated assistant messages in useChat
onFinish not called when stream is aborted
Tool calling with generateObject and streamObject
Abort breaks resumable streams
Streaming Status Shows But No Text Appears
Stale body values with useChat
Unsupported model version error
Object generation failed with OpenAI
Model is not assignable to type "LanguageModelV1"
TypeScript error "Cannot find namespace 'JSX'"
React error "Maximum update depth exceeded"
Jest: cannot find module '@ai-sdk/rsc'
Copy markdown
Tool calling with generateObject and streamObject (structured outputs)
============================================================================================================================================================================================================================
You may want to combine tool calling with structured output generation. While generateObject and streamObject are designed specifically for structured outputs, they don't support tool calling.
To use tool calling with structured outputs, use generateText or streamText with the experimental_output option.
Important: When using experimental_output with tool calling, the structured output generation counts as an additional step in the execution flow.
When using experimental_output with tool calling, adjust your stopWhen condition to account for the additional step required for structured output generation:
const result = await generateText({ model: openai('gpt-4o'), experimental_output: Output.object({ schema: z.object({ summary: z.string(), sentiment: z.enum(['positive', 'neutral', 'negative']), }), }), tools: { analyze: tool({ description: 'Analyze data', inputSchema: z.object({ data: z.string(), }), execute: async ({ data }) => { return { result: 'analyzed' }; }), }, }, // Add at least 1 to your intended step count to account for structured output stopWhen: stepCountIs(3), // Now accounts for: tool call + tool result + structured output prompt: 'Analyze the data and provide a summary',});
For more information about using structured outputs with generateText and streamText see Generating Structured Data
.
On this page
Tool calling with generateObject and streamObject (structured outputs)
Deploy and Scale AI Apps with Vercel.
Vercel delivers the infrastructure and developer experience you need to ship reliable AI-powered applications at scale.
Trusted by industry leaders: