2 mentions in 7 days · 2 in 30 days · 2 total · first seen · last
Z.ai’s GLM-Legal reportedly beat Claude Fable 5 and Kimi K3 on Harvey LAB.
A MarkTechPost headline says Z.ai’s GLM-Legal beat Claude Fable 5 and Kimi K3 on Harvey LAB, a legal-domain AI benchmark. The retrieved page body is a cookie-consent notice, so scores, model sizes, and test details are not available. The item is classified from the headline as a benchmark comparison rather than a confirmed model launch.
Z.ai unveiled GLM-Legal, a GLM-5.3-Flash legal agent that matches frontier models on Harvey, LegalBench, and PLawBench.
Z.ai announced GLM-Legal, a domain model built on GLM-5.3-Flash for agentic legal work such as contract review, case preparation, and document drafting. On Harvey LAB it scored a 19.2% all-pass rate across 24 practice areas, 12.5 points above its base model, and the company says it matches frontier closed models on LegalBench and Chinese PLawBench with a leaner parameter budget. Training combines expanded legal mid-training with reinforcement learning over tens of thousands of legal environments, co-evolved with an agent harness. The model is planned for AutoClaw, Z.ai's desktop agent platform.