Description
Examine reasoning-oriented language models and the methods used to develop their problem-solving behaviour. The course compares them with conventional LLM approaches through tasks such as programming and structured problem solving. DeepSeek-R1 serves as a case study for exploring inference-time scaling, reinforcement learning and knowledge distillation.
The practical discussion considers how model quality, computational efficiency and development cost influence the choice of approach. Exercises use GitHub Codespaces as the development environment described in the course. The focus is on understanding the underlying methods and their tradeoffs so that learners can assess how reasoning-oriented models might fit an application.
Intended audience: AI developers and engineers investigating reasoning models. Researchers and technically experienced learners studying language-model methods.
What you will learn
- Compare reasoning-oriented models with other language-model approaches.
- Explain the roles of inference-time scaling, reinforcement learning and distillation.
- Analyze quality, efficiency and development-cost tradeoffs.
- Explore the course examples in a GitHub Codespaces environment.
Course specifications
- Publisher
- Instructors
- Sebastian Raschka
- Level
- Advanced
- Duration
- 1 hour and 32 minutes
- Subtitle
- English
- Quality
- 1080p
- Edition
- 2026-9
Course topics
View course topics
The course topics image is temporarily unavailable.
Pictures
This picture is temporarily unavailable.
Introduction video
How to open the files
Keep all parts of the same edition together before extracting the first archive. Open the included course media with a compatible player.
Download resources
Choose one edition and keep its parts together. Alternate providers are available when listed.
Size
300 MB
Leave a Reply
Signed-in member comments appear immediately.