GLM 4.6 - What We Learned From 100 Million Open Source Downloads

GLM 4.6 - What We Learned From 100 Million Open Source Downloads

AI Engineer via YouTube Direct link

07:54 - Introducing SLIME: Efficient RL for Agents

5 of 11

5 of 11

07:54 - Introducing SLIME: Efficient RL for Agents

Class Central Classrooms beta

YouTube videos curated by Class Central.

Classroom Contents

GLM 4.6 - What We Learned From 100 Million Open Source Downloads

Automatically move to the next video in the Classroom when playback concludes

  1. 1 0:00 - Introduction & The GLM Ecosystem
  2. 2 0:55 - 100 Million Downloads & Open Source Roadmap
  3. 3 03:22 - Tying GPT-4o on LMSYS Arena
  4. 4 05:04 - The Training Pipeline: From Pre-training to Long Context
  5. 5 07:54 - Introducing SLIME: Efficient RL for Agents
  6. 6 11:08 - The "Two-Stage" Curriculum Strategy
  7. 7 11:57 - Why Single-Stage RL beats Multi-Stage RL
  8. 8 12:55 - Token-Weighted Loss for Coding
  9. 9 14:13 - GLM 4.5V: Multimodal & Video Understanding
  10. 10 16:07 - Deployment: vLLM, SGLang, and Hugging Face
  11. 11 18:06 - Coding Assistants & Future Plans

Never Stop Learning.

Get personalized course recommendations, track subjects and courses with reminders, and more.

Someone learning on their laptop while sitting on the floor.