VeniceAi icon

VeniceAi

Use VeniceAi AI models.

Actions4

Overview

The VeniceAi node's "Chat" operation enables you to interact with Venice AI's chat models, sending user messages and receiving AI-generated responses. It supports advanced features such as system prompts, web search control, memory integration (for conversational context), tool usage (function calling), and the ability to include images in your queries. This node is ideal for building conversational AI workflows, virtual assistants, customer support bots, or any scenario where dynamic, context-aware text generation is needed.

Practical examples:

  • Building a chatbot that can answer questions, use external tools (like calculators or APIs), and remember previous conversations.
  • Creating an AI assistant that analyzes uploaded images and provides descriptions or answers about them.
  • Integrating with other n8n nodes to automate complex, multi-step tasks using natural language.

Properties

Name Type Meaning
Model Name or ID options Selects the Venice AI model to use for chat. You can choose from available models or specify an ID via expression.
Message string The user's message to send to the chat model.
Temperature number Controls randomness of the output; higher values produce more creative responses.
Chat Options collection Additional settings for the chat request (see below for sub-options).
└ System Prompt string Sets the behavior of the assistant by providing a system-level instruction.
└ Enable Web Search options Controls if/when the AI should search the web ("Auto", "Always", "Never").
└ Frequency Penalty number Penalizes new tokens based on frequency; range -2.0 to 2.0.
└ Max Tokens number Maximum number of tokens to generate in the response.
└ Presence Penalty number Penalizes new tokens based on whether they appear in the text so far; range -2.0 to 2.0.
└ Top P number Nucleus sampling parameter; alternative to temperature.
└ Include Venice System Prompt boolean Whether to include the default Venice system prompt.
└ Disable Tools boolean Disables tool/function calling for this request.
Binary Image boolean If enabled, includes an image from binary input as part of the chat message.
Binary Image Property string Name of the binary property containing the image to include (only shown if Binary Image is true).

Output

The node outputs an object with the following structure in the json field:

{
  "output": "<AI response text>",
  "messages": [ /* array of message objects exchanged in the conversation */ ],
  "debug_info": {
    "memory": {
      "connected": <boolean>,
      "available": <boolean>,
      "data_length": <number>,
      "type": "<string>"
    },
    "tools": {
      "connected": <boolean>,
      "count": <number>,
      "names": [ "<toolName1>", "<toolName2>", ... ]
    }
  }
}
  • output: The main AI-generated response as a string.
  • messages: An array representing the full message history sent to the model, including system/user/assistant/tool messages.
  • debug_info: Diagnostic information about memory and tool connections for troubleshooting and transparency.

If the node is configured to process binary data (e.g., when including an image), the output may also contain a binary field with the processed image, but for the "Chat" operation, the primary output is textual.

Dependencies

  • External Service: Requires access to the Venice AI API.
  • API Key: Must be provided via n8n credentials (veniceAiApi), including:
    • apiKey: Your Venice AI API key.
    • baseUrl: Base URL for the Venice AI API (default: https://api.venice.ai).
  • n8n Configuration:
    • For memory/contextual chat, connect an "AI Memory" node.
    • For tool/function calling, connect one or more "AI Tool" nodes.

Troubleshooting

Common Issues:

  • Missing API Key:
    Error: "No valid API key provided"
    Resolution: Ensure your Venice AI credentials are set up correctly in n8n.

  • Invalid Response Format:
    Error: "Invalid response format from Venice AI API. Check your node configuration, especially the system prompt settings."
    Resolution: Double-check your model selection, system prompt, and ensure the API is returning expected data.

  • No Models Found:
    Error: "No models found for operation: chat"
    Resolution: Verify your API key and base URL, and that your account has access to chat models.

  • Binary Data Errors:
    Error: "No binary data exists on item!" or "No binary data property '...' exists on item!"
    Resolution: Make sure the upstream node provides the required binary data under the correct property name.

  • Tool Call Errors:
    Error: "Tool does not have a valid call() or _call() method"
    Resolution: Ensure connected tool nodes implement the required methods.

  • Unexpected Output:
    If the output is not as expected, check the debug_info section for clues about memory and tool connectivity.

Links and References

Discussion