Prompt DeepSeek's R1
This workflow shows how to connect to DeepSeek's reasoning model, R-1, and send prompts to it.
In the first case, the connection is established via DeepSeek's API using an API key. If you don't have one already, register with DeepSeek and create a new API key under: https://platform.deepseek.com/api_keys
In the second case, the model is sourced from Ollama, and the connection happens locally. To set up Ollama and leverage its open-source models locally in KNIME, follow these steps: https://www.knime.com/blog/how-to-leverage-open-source-llms-ollama.