DeepSeek icon

DeepSeek

Consume DeepSeek AI

Actions2

Overview

The DeepSeek node for n8n enables users to interact with DeepSeek AI's chat models. It allows you to send a series of messages (as a conversation) to an AI model and receive generated responses, making it suitable for building conversational bots, automating customer support, or integrating advanced language understanding into workflows.

Common scenarios:

  • Automating Q&A or helpdesk tasks.
  • Generating creative content or summaries based on user prompts.
  • Building interactive assistants that require context-aware responses.

Example:
You can use this node to pass a sequence of user and system messages to a DeepSeek model and get a relevant AI-generated reply, which can then be used in subsequent workflow steps.

Properties

Below are the input properties supported by the node for the Chat resource and its default operation:

Display Name Type Description
Model options The model which will generate the completion. Learn more.
Prompt fixedCollection A collection of messages forming the conversation. Each message includes:
- Role (options: Assistant, System, User): The role of the message sender.
- Content (string): The actual message text.
Simplify boolean Whether to return a simplified version of the response instead of the raw data.
Options collection Additional parameters to control the model's output:
- Frequency Penalty number Penalizes new tokens based on their frequency in the text so far.
- Maximum Number of Tokens number The maximum number of tokens to generate in the completion.
- Presence Penalty number Penalizes new tokens based on whether they appear in the text so far.
- Sampling Temperature number Controls randomness; lower values make output more deterministic.
- Top P number Controls diversity via nucleus sampling.
- Response Format json Specifies the format that the model must output.
- Logprobs boolean Whether to return log probabilities of the output tokens.
- Top Logprobs number Number of most likely tokens to return at each token position (requires Logprobs to be true).

Output

  • json:

    • If Simplify is enabled (default):
      The output will be a simplified array of choices returned by the model, typically containing the generated message(s).
      [
  {
    "message": {
      "role": "assistant",
      "content": "AI-generated response here"
    },
    ...
  }
]
    • If Simplify is disabled:
      The full raw response from the DeepSeek API is returned, which may include additional metadata such as usage statistics, model information, etc.

  • Binary Data:
    This node does not output binary data.

Dependencies

  • External Service: Requires access to the DeepSeek API.
  • API Key: You must configure valid DeepSeek API credentials in n8n under the name deepSeekApi.
  • n8n Configuration: No special environment variables required beyond standard credential setup.

Troubleshooting

Common Issues:

  • Invalid API Key:
    Error if the provided API key is missing or incorrect. Ensure your credentials are set up correctly in n8n.
  • Model Not Selected:
    If no model is chosen, the request may fail. Always select a valid model from the dropdown.
  • Malformed Prompt:
    If the prompt/messages structure is invalid (e.g., missing roles or content), the API may reject the request.
  • Token Limits Exceeded:
    If you request more tokens than allowed by the selected model, you'll receive an error. Adjust the "Maximum Number of Tokens" accordingly.

Error Messages & Resolutions:

  • "401 Unauthorized": Check your API credentials.
  • "400 Bad Request": Review your input fields, especially the prompt/message structure and options.
  • "429 Too Many Requests": You have hit rate limits; try again later or reduce request frequency.

Links and References

Discussion