> ## Documentation Index
> Fetch the complete documentation index at: https://orangeslice.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Generate Object

> Generate structured JSON from natural language

<Info>
  <Icon icon="coins" /> **1 credit** per call
</Info>

## Method

```typescript theme={null}
services.ai.generateObject(params);
```

<Note>
  Does NOT have internet access. Use for extraction/parsing, not research.
</Note>

## Input

<ParamField path="prompt" type="string" required>
  Prompt describing what to generate or extract
</ParamField>

<ParamField path="schema" type="ZodSchema" required>
  Zod schema defining output structure (must be `z.object` at top level)
</ParamField>

<ParamField path="model" type="string" default="gpt-5-mini">
  Model: `"gpt-5-mini"`, `"gemini-2.0-flash"`, `"gemini-2.5-pro"`, or
  `"claude-sonnet-4-5-20250929"`
</ParamField>

## Output

<ResponseField name="object" type="object">
  Generated JSON matching the provided schema
</ResponseField>

## Example

```typescript theme={null}
const result = await services.ai.generateObject({
  prompt: `Extract company info: ${ctx.thisRow.get("Description")}`,
  schema: z.object({
    industry: z.string(),
    products: z.array(z.string()),
    target_market: z.enum(["B2B", "B2C", "Both"]),
  }),
});

return result.object.industry;
```