Z.ai releases GLM-5.3-FlashX-Legal: 1M-token legal model priced at $0.37/$1.25 per 1M tokens

Z.ai released GLM-5.3-FlashX-Legal (GLM-Legal) on November 2, 2026, a legal-domain variant of GLM-5.3-FlashX with a 1M-token context window. It is listed on OpenRouter at $0.37 per 1M input tokens and $1.25 per 1M output tokens.

Key specifications

• Model: GLM-5.3-FlashX-Legal (z-ai/glm-5.3-flashx-legal)
• Input / output: text, image and video in; text out
• Context window: 1.0M tokens
• Pricing: $0.37 input / $1.25 output per 1M tokens; cache read $0.09 per 1M tokens
• Thinking mode: mandatory
• Supported features: function calling, structured output, streaming
• Provider: Z.ai (the only provider listed on OpenRouter)
• Release date: November 2, 2026
• Regional availability: not currently available through EU in-region routing

What Z.ai changed

According to the model description, GLM-Legal is built on GLM-5.3-FlashX, a native multimodal model. Z.ai adds legal-domain mid-training and post-training with reinforcement learning in task environments modeled on real legal workflows.

The stated targets are document understanding, legal reasoning, and multi-step task execution. The listed tasks are legal research, contract review, cross-document consistency checks, and drafting or revising legal documents.

The model keeps the GLM-5.3-FlashX 1M-token context window, mandatory thinking mode, function calling, structured output, and streaming. It is designed to run in agent setups connected to legal data and document-editing tools through MCP services.

The parent model

GLM-5.3-FlashX is described as the high-speed variant of Z.ai's GLM-5.3-Flash, with inference speeds of up to 200 tokens/s. It uses a hybrid sparse and linear attention architecture with 320B total parameters and 18B active, and is suited to efficient coding, visual understanding, and long-horizon agent tasks. GLM-Legal's price matches the parent's listed price of $0.37 / $1.25 per 1M tokens.

Other GLM-5.3 family models on the same page include GLM-5.3 ($0.0699 / $1.40), GLM-5.3 Prime ($2.80 / $8.80), and GLM-5.3 Flash ($0.0285 / $0.0923), all with 1M-token context.

Not yet disclosed

Benchmark scores for GLM-Legal are not yet disclosed. Latency and throughput figures were not yet populated on the OpenRouter listing as of November 2, 2026, and the page showed too little uptime data to display a history.

What this means

GLM-Legal is a domain-specific variant, not a general-purpose release: the legal training and RL environments are aimed at workflows such as contract review and cross-document consistency checks. The 1M-token context window is relevant to those tasks, which involve comparing multiple long documents.

The design for MCP-connected agent setups means the model is intended to work with legal data sources and document-editing tools rather than as a standalone chat model. Teams evaluating it will need their own testing, since no benchmark results are disclosed. The price is identical to the general GLM-5.3-FlashX, so the listing shows no price premium for the legal specialization. The model is not available through EU in-region routing, which matters for teams with regional data-handling requirements.