Full video course on building AI reasoning models from scratch
In brief
- 11-hour video course to implement reasoning models in Python and PyTorch.
- Uses a pre-trained LLM (Qwen3 0.6B) and adds evaluation and optimization methods.
- Covers inference, real-time scaling, reinforcement learning, and distillation.
Sebastian Raschka introduces an 11-hour video course to implement reasoning models in Python and PyTorch. According to the author, these models, combining reinforcement learning and scaling techniques, are central to today’s LLMs.
The course details evaluation, optimization, and adding reasoning capabilities to a pre-trained LLM (Qwen3 0.6B). It covers inference, real-time scaling, reinforcement learning with verifiable rewards, and distillation. Videos are paired with a companion book and include hands-on exercises.
These resources aim to demystify the inner workings of reasoning models, often used as engines in AI agents. The author recommends proficiency in Python and PyTorch, with a GPU for advanced experiments.
This summary gives you the essentials. The full article, its figures and quotes are on Ahead of AI.
Read the original articleAI-generated summary of the source, checked by rules. How we write · Report an error