Z.AI’s developer documentation says GLM Coding Plan users on Max, Pro, and Lite tiers can switch to GLM-5.2, including a glm-5.2[1m] coding-agent configuration with a 1,000,000-token context window and 131,072 max tokens.
Z.AI has updated its developer documentation to say that GLM Coding Plan now supports GLM-5.2 for Max, Pro, and Lite users.
In a developer documentation page titled “How to Switch Models,” Z.AI states that GLM Coding Plan users on the Max, Pro, and Lite plans can use GLM-5.2. The same Z.AI documentation describes a glm-5.2[1m] model option for coding-agent configurations.
According to the Z.AI developer docs, that glm-5.2[1m] configuration has a 1,000,000-token context window and supports up to 131,072 max tokens in coding-agent settings. The documentation frames the page as guidance for switching models, rather than as a broader product announcement.
Z.AI’s developer documentation also includes a separate “How to Switch Models” page for GLM-5.1. That page says GLM Coding Plan supports GLM-5.1 for Max, Pro, and Lite users, and provides configuration steps for tools including Claude Code, OpenClaw, and Cline.
The GLM-5.1 documentation is useful context because it shows Z.AI has been using its developer docs to explain how coding-plan customers can select newer GLM models inside developer tools. The newer GLM-5.2 page appears to continue that pattern by documenting model-switching details for the next model version.
Z.AI’s “New Released” release-notes page lists GLM-5.1 on 2026-04-07. In that release-note entry, Z.AI describes GLM-5.1 as designed for long-horizon tasks, with stronger engineering intelligence and tool use.
The provided Z.AI sources do not include an equivalent release-note excerpt for GLM-5.2. Based on the available documentation, the confirmed update is narrower: Z.AI’s developer docs say GLM Coding Plan supports GLM-5.2 and document a glm-5.2[1m] coding-agent configuration with a one-million-token context window.
For developers using coding agents, context-window size can affect how much code, documentation, and task history a model can consider at once. Z.AI’s documented 1,000,000-token context window for glm-5.2[1m] suggests the configuration is aimed at workflows involving large repositories or extended coding sessions.
The 131,072 max-token setting cited in Z.AI’s documentation is also relevant for coding-agent use, because it defines the maximum output-token configuration described for that model setup. The source excerpt does not specify performance benchmarks, pricing details, rate limits, or availability outside Max, Pro, and Lite GLM Coding Plan users.
Z.AI’s developer documentation now identifies GLM-5.2 as supported for GLM Coding Plan users on Max, Pro, and Lite tiers. The key documented configuration is glm-5.2[1m], which Z.AI says uses a 1,000,000-token context window and 131,072 max tokens for coding-agent setups.
The update is best read as a developer-documentation change focused on model switching and configuration. Z.AI’s provided docs do not support broader claims about real-world performance, benchmark leadership, or general availability beyond the plan tiers and coding-agent configuration details stated in the documentation.
Z.AI has updated its developer documentation to say that GLM Coding Plan now supports GLM 5.2 for Max, Pro, and Lite users.
What changed In a developer documentation page titled “How to Switch Models,” Z.AI states that GLM Coding Plan users on the Max, Pro, and Lite plans can use GLM 5.2.
The same Z.AI documentation describes a glm 5.2[1m] model option for coding agent configurations.
Continue reading
