GLM-4.5 Team: integrated agentic, reasoning and coding model training

What changed

Combined staged pre-training, expert-model iteration and reinforcement learning in a hybrid thinking/direct-response expert-model family.

Scientific scope

Combined staged pre-training, expert-model iteration and reinforcement learning in a hybrid thinking/direct-response expert-model family. One integrated report; hybrid reasoning and mixture-of-experts are not invented here. Rankings are tied to the July 2025 evaluation setup, not current or universal superiority.

Evidence

The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. One integrated report; hybrid reasoning and mixture-of-experts are not invented here. Rankings are tied to the July 2025 evaluation setup, not current or universal superiority.

Contributors

• Source-listed authors and technical contributors · SCIENCE: Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.

• GLM-4.5: Agentic, Reasoning, and Coding (ARC) Foundation Models
• GLM-4.5 Team: integrated agentic, reasoning and coding model training — official team research record

Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.

Full scientific record · Browse contribution grades · Request a correction