PrompteryxBeta
    ⌘K
    PlatformSupport
    Menu
    IntroductionInstall Desktop & extensionQuick StartPlatform Overview
    OverviewWhat you can makeAI Browser Agent in the chatModels and AutoConnectorsSchedulesProjectsExplore and publishingShare linksMemory and SkillsPersonal agent (beta)Welcome, settings and skinsPlans, credits and top-upsPrompteryx DesktopGroups (pilot)Phone app (early access)
    OverviewCreating WorkflowsTesting & DebuggingScheduling
    OverviewGetting StartedCommandsRecordingSettingsStructured OutputParallel AgentsTips & Best Practices
    OverviewInstallationSidebar AgentUsing Attached FilesExecute on Current TabExporting to Visual Studio
    Overview
    OverviewOpen PageClickClick CoordinatesFillExtract TextScreenshotWaitPress KeyFocus ElementBlur ElementSelect TextHighlight ElementSet ViewportGet Bounding BoxScrollHoverSelect OptionAssertionsCookiesEvaluate JavaScriptExample Workflows
    OverviewPrompteryx Browser AgentOpenAI
    OverviewSchedule TriggerWebhook Trigger
    OverviewGoogle SheetsGoogle CalendarConnect Hub (3,000+ apps)Example Workflows
    AI Vision SettingsChrome Automation ProfilesCloud BrowserBrowser Profiles & Persistent ContextVideo RecordingExecution LogsSharing & CollaborationEdit History
    Selectors GuideVariables
    Start hereGet an API keyRun the AI browser agentDrive a cloud browserRun workflows from anywhereUse connected apps (Connect Hub)SDK (TypeScript)CLIMCP for AI assistants
    REST APIWebhooksMCP toolsCloud Browser APICloud Browser quickstartMigrate from Browserbasen8nZapierMake.com

    Need help? Contact Support

    1. Documentation
    3. Nodes
    5. Integrations
    7. OpenAI

    OpenAI / ChatGPT

    AI Integration

    Generate text, images, and audio with OpenAI models

    Overview

    The OpenAI integration connects your workflows to OpenAI's GPT models, GPT Image models, and audio models. Use it to generate text, summarize content, analyze data, create and analyze images, and add intelligence to your automations.

    Operations are grouped by resource: Text (Message a Model, Classify Text for Violations), Image (Analyze Image, Generate an Image), Audio (Generate Audio, Transcribe a Recording, Translate a Recording) and File (Upload a File, List Files, Delete a File). The most common ones are below.

    Available Operations

    Message a Model
    Generate text responses using GPT models (Text resource)
    Parameters
    Configuration options for this node
    Model
    Select
    Required

    The GPT model to use

    Default: gpt-5.5
    Example: gpt-6-astra, gpt-5.5, gpt-5.4-mini, gpt-4o
    Prompt
    String
    Required

    The prompt to send to the model

    Example: Summarize this article in 3 bullet points: {{articleText}}
    Role
    Select

    User (ask and get a response), System (set behavior or context) or Assistant (set a tone or personality)

    Default: User
    Simplify Output
    Boolean

    Return a simplified version of the response instead of the raw data

    Default: true
    Output Content as JSON
    Boolean

    Ask the model to return its response as JSON

    Default: false
    Options → Temperature
    Number

    Randomness (0-2). Lower = more focused and deterministic. Options also include Maximum Number of Tokens (default 4096), Top P, and Frequency / Presence Penalty.

    Default: 1
    Analyze Image
    Ask a vision-capable model about an image (Image resource)
    Parameters
    Configuration options for this node
    Model
    Select
    Required

    Vision-capable model used to look at the image

    Default: gpt-4o
    Example: gpt-6-astra, gpt-5.5, gpt-4.1, gpt-4o-mini
    Text Input
    String
    Required

    The question or prompt about the image

    Default: What's in this image?
    Input Type
    Select

    Image URL(s) or Binary File(s) (a data: URI or base64 from an earlier node)

    Default: Image URL(s)
    Image URL(s)
    String

    A hosted image URL (several can be separated by commas), or a data: URI from an earlier node

    Example: https://example.com/image.png
    Options → Max Tokens
    Number

    Maximum length of the answer. Options also include Detail Level (Auto, Low, High).

    Default: 300
    Generate an Image
    Create images from text descriptions (Image resource)
    Parameters
    Configuration options for this node
    Model
    Select
    Required

    GPT Image model

    Default: gpt-image-2
    Example: gpt-image-2.6-sunburst, gpt-image-2.6-flare, gpt-image-2.5-flare, gpt-image-2, gpt-image-1-mini
    Prompt
    String
    Required

    Description of the image to generate

    Example: A professional photo of a modern office workspace
    Options → Resolution
    Select

    Image dimensions: Auto, 1024x1024 (square), 1536x1024 (landscape) or 1024x1536 (portrait)

    Default: 1024x1024
    Options → Quality
    Select

    Auto, Low (fastest, cheapest), Medium or High

    Default: Auto
    Options → Return Image URL
    Boolean

    GPT Image returns base64 only; when on, the output's url field carries the image as a data URI so later steps can use it

    Default: true

    Usage Examples

    Example 1: Summarize Web Page
    Extract text and generate a summary

    Workflow:

    1. Open Page: Article URL
    2. Extract Text: Article content → articleText
    3. OpenAI — Message a Model:
    4. Prompt: "Summarize this article in 3 bullet points: {{articleText}}"
    5. Send summary via Slack/Email
    Example 2: Classify Support Tickets
    Automatically categorize incoming tickets

    Workflow:

    1. Webhook Trigger: Receive new ticket
    2. OpenAI — Message a Model:
    3. Prompt: "Classify this ticket as billing, technical, feature-request or other: [ticket text]" — insert the ticket text from the trigger with the {x} variable button
    4. Route ticket based on classification
    Example 3: Generate Product Descriptions
    Create marketing copy from product data

    Workflow:

    1. Google Sheets Read: Product data
    2. For Each: Process each product
    3. ↳ OpenAI — Message a Model: Generate description
    4. ↳ Update Sheet with generated text

    Model Comparison

    GPT-6 Astra (Latest)

    The newest model in the list — OpenAI's current flagship.

    GPT-6.1 Terra

    OpenAI's newest value model (October 2026): strong results at a low price, and it reads images too.

    GPT-5.5

    The default for Message a Model. A good starting point.

    Mini and nano models

    Smaller variants such as GPT-5.4 mini and GPT-5.4 nano. The list also includes older models (GPT-4.1, GPT-4o, o3 and more).

    Best Practices

    Write Clear Prompts
    Be specific about the format you want, the tone, and any constraints. Example: "Respond in JSON format with keys: category, confidence, reason" — and turn on Output Content as JSON.
    Use Lower Temperature for Consistency
    For classification, extraction, or structured output, use temperature 0-0.3. Higher temperatures (0.7-1.0) are better for creative content.
    Set Appropriate Max Tokens
    The Maximum Number of Tokens option defaults to 4096. Lower it when you expect short answers — lower values save costs and reduce response time.
    API Key Required
    You'll need an OpenAI API key. Get one from platform.openai.com/api-keys and add it in the node's “Credential to connect with” field (Create new credential) or under Connections → Keys. Usage is billed by OpenAI separately.

    Related Documentation

    AI Browser Agent

    AI-powered browser automation

    Variables Guide

    Use AI responses in workflows