88Z AI, 13 October — Zhipu has released GLM-Legal, a legal-specific large model built on the GLM-5.3-Flash base. It covers the entire workflow from contract review to document generation. The model achieved a 19.2% pass rate in Harvey LAB tests, matching leading closed-source models, and supports API and AutoClaw desktop access with a focus on low cost and traceable evidence.
At a glance
- Built on GLM-5.3-Flash with a 19.2% Harvey LAB pass rate, matching leading closed-source models
- Covers full legal workflows via API and AutoClaw desktop, emphasizing low cost and traceable evidence
The update
Zhipu has officially launched GLM-Legal, a large model dedicated to the legal sector. Built on the lightweight GLM-5.3-Flash base, it has been optimized through specialized legal training and reinforcement learning. Its capabilities cover the entire workflow, including contract review, evidence analysis, legal research, compliance planning, and document generation. Official data indicates a 19.2% overall pass rate in Harvey LAB benchmarks, with performance on LegalBench and PLawBench matching that of leading closed-source models. For deployment, GLM-Legal supports both API access and the AutoClaw desktop application. The product emphasizes low costs, traceability of evidence, and the immediate usability of outputs, aiming to enhance efficiency and standardization in legal work.
Sources & editorial note
This article is an AI-assisted synthesis by 88Z AI based on the public sources below. Product claims and availability are attributed to their publishers. It does not represent independent testing or first-hand reporting.
- GLM-LegalSource published
Source dates above refer to the original announcements; the date under the headline is this article’s publication on 88Z AI.