Post training for AI models stories

Aug 14, 2026

Z.ai releases GLM-5.3 model with improved coding capabilities

Chinese AI company Z.ai released GLM-5.3, upgrading its flagship model through scaled post-training without retraining the base model. The model shows improvements in complex coding and long-horizon tasks.

Aug 14, 20263 sources

May 10, 2026

NVIDIA Releases Star Elastic: Single Checkpoint Contains Multiple Reasoning Models

NVIDIA researchers have introduced Star Elastic, a post-training method that embeds multiple nested reasoning models at 30B, 23B, and 12B parameter scales inside a single checkpoint, enabling zero-shot model slicing without retraining.

May 10, 20261 source

Top claims

▪The voluntary two-week delay of GLM-5.3 marks the first time a Chinese frontier AI lab has restricted open-weight release timing due to specific emergent capability concerns
▪The oldest vulnerability discovered by Z.ai's models in real deployed software dates back to 1981
▪Z.ai scaled GLM-5.3's post-training by increasing the volume of long-horizon task environments, using automated pipelines with research and judge agents to convert real work patterns into solvable tasks

Subtopics

AI foundation models2 AI coding assistants1 AI research & benchmarks1 China tech & industrial strategy1 Large language models (LLMs)1 Reasoning models1

Related timelines