Initial thoughts on DeepSeek

I recently started using the DeepSeek language model. I gave in because it is “open-source” (in terms of LLMs) and because there’s been a lot of talk about it in recent days. Initially I installed ollama in order to be able to pull the DeepSeek R1 model and use it via TTY or CLI. This is convenient if you wish to access it via SSH on a less capable machine such as a laptop. It is also possible to access via a GUI such as chatbox.

The specific condensed model which I chose is the 8B version, which has 8 billion parameters. Once ollama is installed, the 8B version of the DeepSeek R1 model can be pulled by running the following command: ollama run deepseek-r1:8b This command can then be used to start interacting with the chatbot: ollama run deepseek-r1:8b