Skip to content
NEWError doctor and Watchtower: failed n8n runs explained in plain words, and emailed to you. See what is new

Library / AI Agents & Assistants

AI Agents & Assistants

Chat with a DeepSeek Telegram Agent with Long-Term Memory

When a Telegram message arrives from an allowed user, a DeepSeek agent replies using window memory and long-term memories saved in Google Docs.

What it does

# DeepSeek API Call The DeepSeek API uses an API format compatible with OpenAI. By modifying the configuration, you can use the OpenAI SDK or softwares compatible with the OpenAI API to access the DeepSeek API. {{ $env.API_BASE_URL }} ## Configuration Parameters | Parameter | Value | |-----------|--------| | base_url | {{ $env.API_BASE_URL }} | | api_key | {{ $env.API_BASE_URL }} | ## Important Notes - To be compatible with OpenAI, you can also use {{ $env.API_BASE_URL }} as the base_url. Note that the v1 here has NO relationship with the model's version. - The deepseek-chat model has been upgraded to DeepSeek-V3. The API remains unchanged. You can invoke DeepSeek-V3 by specifying model='deepseek-chat'. - deepseek-reasoner is the latest reasoning model, DeepSeek-R1, released by DeepSeek. You can invoke DeepSeek-R1 by specifying model='deepseek-reasoner'.

The steps

  1. Check User & Chat ID
  2. Error message
  3. Listen for Telegram Events
  4. Validation
  5. Message Router
  6. AI Agent
  7. Merge
  8. Window Buffer Memory
  9. When chat message received
  10. Telegram Response
  11. Save Long Term Memories
  12. Retrieve Long Term Memories
  13. Response Error message
  14. DeepSeek-R1 Reasoning

32 steps in total.

Similar automations