Resources
Course · MIT Han Lab · Fall 2026 · 19 lectures · 21h
6.5940: TinyML and Efficient AI Computing
Song Han
Making models cheap to run, from pruning, quantization and distillation to distributed training, serving and acceleration for LLMs and diffusion models.
Efficient AIML systemsLLMs
- Step 2 of 3 in LLMs, from the Transformer to the data center
0 of 19 watched21h left
- 1h 3m
- 58m
- 42m
- 1h 14m
- 1h 8m
- 1h 2m
- 1h 1m
- 1h 9m
- 59m
- 1h 3m
- 1h 16m
- 1h 17m
- 1h 9m
- 1h 16m
- 1h 9m
- 1h 18m
- 1h 6m
- 1h 6m
- 53m
Your progress is saved in this browser only. Back it up to keep it or move it to another device.
Course materials
Published by the course. These open on its own site.