Skip to main content

GPT Structured Output

Force the model to return data in a specific JSON format. Define your output structure using a JSON Schema — the model guarantees its response matches it exactly.

The Structured Output node uses an LLM to return data that follows a schema you define, instead of free-form text. Describe the fields you expect and the node produces clean, structured JSON for the rest of your flow.

GPT Structured Output
System MessageText
ContextText
Prompt*Text (multiline)
Json Schema*JSON
Image URLFile upload
ModelSelect
Fields marked * are required. Dots are connection handles.

Inputs

Each input can be typed in directly or connected to another node's output.

InputTypeRequiredDescription
System MessageTextOptionalOverride the default system instructions for the LLM (e.g. 'You are a concise assistant that responds only in bullet points.').
ContextTextOptionalDynamic data for the LLM to process (e.g. email body, document, user input).
PromptText (multiline)RequiredTask instruction for the LLM (e.g. 'Extract the name and date from the text').
Json SchemaJSONRequiredJSON Schema defining the exact structure of the output. The model will always return valid JSON matching this schema.
Image URLFile uploadOptionalProvide images for the LLM to analyze by connecting image URLs to this field.

Settings

SettingTypeDefaultDescription
ModelSelectGPT-5.5

Model options

OptionDefault
GPT-4o-mini
GPT-4o
GPT-4.1
GPT-4.1-mini
GPT-5.4
GPT-5.4-mini
GPT-5.4-nano
GPT-5.4-pro
GPT-5.5
GPT-5.5-pro

Output

Type: JSON (json)