DeepSeek-R1 Crash Course
freeCodeCamp.org
Andrew Brown of ExamPro explains DeepSeek-R1, the reasoning model trained with reinforcement learning that performs close to OpenAI’s o1 at a much lower cost. Then he runs it in different ways: on DeepSeek’s website, locally with Ollama and LM Studio, and in code with Hugging Face Transformers.
What you'll learn
- What DeepSeek-R1 is and how it was trained
- Using DeepSeek on its website
- Running DeepSeek-R1 locally with Ollama and LM Studio
- Loading the model with Hugging Face Transformers
