GLM-Legal
A specialized large model and agent for the legal domain, built on the GLM series of models. It supports legal reasoning, tool calling, statute retrieval, case-law lookup and structured output, with emphasis on interpretable and traceable conclusions.
Timeline (last 90 days)
In the VALS AI legal evaluation, GLM-Legal took first place among open-weight models on both the Harvey Lab and Legal Bench leaderboards, with a 19.2% all-tasks-passed rate and 86.23% accuracy.
Zhipu launched GLM-Legal, an open-source large model dedicated to law, for practical work such as contract review, evidence linking and cross-border compliance research.
GLM-Legal is built on the base model GLM-5.3-Flash, belonging to the 'base + domain fine-tuning' route.
GLM-Legal's score on Harvey LAB is close to three times that of its base model GLM-5.3-Flash.
The GLM-Legal model launched by Z.ai achieved a 19.2% all-pass rate on the Harvey LAB legal Agent benchmark, surpassing Claude Fable 5 and Kimi K3.
The running cost of each test case on Harvey LAB is about $0.15.
Currently, access to GLM-Legal is by application, rather than a directly open API.
GLM-Legal's 19.2% all-pass rate on Harvey LAB means that on most tasks the model still needs a human lawyer as a fallback.
GLM-Legal is positioned as 'Agentic Legal Intelligence,' that is, a legal agent emphasizing autonomous action capabilities, rather than a simple legal question-answering bot.
The design of GLM-Legal enables the model to proactively retrieve legal provisions, query case law, and call calculation tools to estimate compensation amounts during reasoning.
1 more timeline events
All Facts (11)
In the VALS AI legal evaluation, GLM-Legal took first place among open-weight models on both the Harvey Lab and Legal Bench leaderboards, with a 19.2% all-tasks-passed rate and 86.23% accuracy.
50%UnverifiedGLM-Legal's score on Harvey LAB is close to three times that of its base model GLM-5.3-Flash.
50%UnverifiedGLM-Legal is built on the base model GLM-5.3-Flash, belonging to the 'base + domain fine-tuning' route.
50%UnverifiedThe running cost of each test case on Harvey LAB is about $0.15.
50%UnverifiedCurrently, access to GLM-Legal is by application, rather than a directly open API.
50%UnverifiedGLM-Legal's 19.2% all-pass rate on Harvey LAB means that on most tasks the model still needs a human lawyer as a fallback.
50%UnverifiedThe GLM-Legal model launched by Z.ai achieved a 19.2% all-pass rate on the Harvey LAB legal Agent benchmark, surpassing Claude Fable 5 and Kimi K3.
50%UnverifiedGLM-Legal is said to have made specialized optimizations in legal corpus, reasoning chains, and output format.
50%UnverifiedThe design of GLM-Legal enables the model to proactively retrieve legal provisions, query case law, and call calculation tools to estimate compensation amounts during reasoning.
50%UnverifiedGLM-Legal is positioned as 'Agentic Legal Intelligence,' that is, a legal agent emphasizing autonomous action capabilities, rather than a simple legal question-answering bot.
50%UnverifiedZhipu launched GLM-Legal, an open-source large model dedicated to law, for practical work such as contract review, evidence linking and cross-border compliance research.
50%