Zhipu's Tang Jie Reveals GLM-5.5 Training Pipeline at Tsinghua Talk

Zhipu chief scientist Tang Jie revealed the training pipeline of the upcoming GLM-5.5 at a Tsinghua lecture, with sizes of 32B, 122B, and 256B. Its pretraining relies not on public datasets but on six internal data engines, drawing excitement from the community.

2026-11-05 ~ 2026-11-06 · 3 related posts

• Zhipu chief scientist reveals GLM-5.5 training pipeline: 1T web pages, six data engines — algo_diver · 2026-11-05
• Z.ai's chief scientist reveals GLM-5.5 pipeline: 32B, 122B, and 256B sizes coming — wapswaps · 2026-11-06
• Zhipu chief scientist reveals GLM-5.5 training pipeline: 3 sizes, 6 internal data engines — AccBalanced · 2026-11-06