Overview
This node integrates with the DeepSeek AI platform to generate chat completions using specified AI models. It is designed for scenarios where users want to generate conversational AI responses based on a series of messages with roles such as user, assistant, or system. Practical applications include building chatbots, virtual assistants, or any system requiring AI-generated conversational text.
Use Case Examples
- Generating a customer support chatbot response based on user input messages.
- Creating an AI assistant that can continue a conversation with context from previous messages.
Properties
| Name | Meaning |
|---|---|
| Model | The AI model used to generate the chat completion. Models are dynamically loaded from the DeepSeek API and filtered to those starting with 'deepseek-'. |
| Prompt | A collection of messages forming the conversation history, each with a role (assistant, system, user) and content, which serves as the input prompt for the AI model. |
| Simplify | A boolean flag to return a simplified version of the response containing only the relevant data (choices) instead of the full raw API response. |
| Options | Additional optional parameters to customize the completion generation, such as frequency penalty, presence penalty, maximum tokens, temperature, log probabilities, response format, and sampling controls. |
Output
JSON
data- The main response data containing the AI-generated completion choices.
Dependencies
- DeepSeek API
Troubleshooting
- Ensure the API key credential for DeepSeek is correctly configured and valid to avoid authentication errors.
- If the model list is empty, verify network connectivity and API endpoint accessibility.
- Incorrect or missing prompt messages may lead to unexpected or empty completions; ensure messages are properly structured with roles and content.
- If the simplified output option is enabled but the response format changes, the node might fail to extract the data correctly.
Links
- DeepSeek API Pricing and Models - Official documentation for DeepSeek models and pricing details.