Zhupai (z.ai) has released GLM-5, an open-source 744B parameter model that achieves a record-low hal...
The AMW Read
The release of a 744B open-weight model that matches frontier performance at 1/6th the cost validates the CN/OSS challenger frame (Frame 2) and signals a structural shift in scaling-law economics.
NoveltySignificance
Foundation Models Β· Player MapScaling Laws
Zhupai (z.ai) has released GLM-5, an open-source 744B parameter model that achieves a record-low hallucination score of -1 on the AA-Omniscience Index. By leveraging a novel asynchronous RL infrastructure called slime, it delivers frontier-level performance at 1/6th the cost of proprietary rivals. The model moves beyond chat with a native Agent Mode that autonomously generates professional office documents. This release marks a systemic shift toward reliable, low-cost autonomous engineering for the global enterprise. π
