ZCode 宣布用户达百万，并称将重置部分套餐额度

To make usage clearer and easier to predict, GLM Coding Plan is moving to a credits-based system. See the Subscription page and the usage guides for Individual Plans and Team Plans for new plan tiers, prices, credit amounts, and deduction rules. Key points The new credits-based plan is now available. Previous plans are no longer sold to new users. Existing active subscriptions are not affected. In addition, usage on weekends will be deducted at off-peak rates all day. Users on plans discontinued on July 30 can still renew or upgrade them. Discontinued and new plans differ only in how usage is calculated. Other benefits will remain aligned. ## ​ Identify Your Current Plan Sign in and use the table below to identify your current plan and see how this update affects you. [...] | Plan Type | 5-Hour Limit (Dynamically refreshed; quota resets 5 hours after consumption) | Weekly Limit (Activated upon subscription; resets every 7 days) | --- | Lite Plan | Up to approx. 80 prompts | U

## ZCode Key Features Long-task execution. The agent maintains context across files, terminal output, browser state, Git state, and execution modes throughout a single task. This is where GLM-5.2’s 1-million-token context window becomes practically useful rather than just a marketing number. Five execution modes. ZCode ships with multiple operation modes (cycled with Shift+Tab) that control how autonomously the agent operates. Less autonomous modes require confirmation before sensitive commands; fully autonomous modes can run without interruption. This dial matters, especially after a few high-profile incidents of agents chaining actions without human review. [...] For developers who are spending $100+ a month on AI coding subscriptions and getting frustrated by rate limits, the Coding Plan is worth a serious trial. The 5-day ZCode free trial costs nothing and will tell you more than any benchmark comparison ever will. The catch is that “worth it” is contextual. If you’re working on

> : refers to their scores of full set. # Getting started with GLM-5.2 ## Use GLM-5.2 with GLM Coding Plan Try GLM-5.2 in your favorite coding agents—ZCode, Claude Code, OpenCode, and more. For GLM Coding Plan subscribers: We already rolled out GLM-5.2 to all Coding Plan users. You can enable GLM-5.2 now by updating the model name to `"GLM-5.2"` (or GLM-5.2[1m] in Claude Code to enable 1M context length). You can also choose different thinking effort, High or Max, depending on the task. As our most capable model, GLM-5.2 consumes quota at 3× during peak hours and 2× during off-peak hours. As a limited-time promotion through the end of September, off-peak usage is billed at 1×. (Peak hours are 14:00–18:00 UTC+8 (Beijing Time) daily). [...] Image 1 2026-06-16 · Research # GLM-5.2: Built for Long-Horizon Tasks Image 2Try it at Z.ai Image 3Call it at Z.ai Image 4Z.ai Coding Plan Image 5GitHub Image 6HuggingFace We're introducing GLM-5.2, our latest flagship model for long-horizon

$12.6/ month$18 10,000 Credits / week Rolling access to the latest flagship models and features Supports 20+ agent tools, including ZCode, Claude Code, and more Default data privacy GLM Coding ### Pro Popular Built for day-to-day development on mid-sized repo $56/ month$80 6× Lite usage All Lite benefits Priority access to the latest flagship models and features Includes a curated selection of MCP tools Faster generation speeds GLM Coding ### Max Max Usage Built for advanced users working on mid-to-large repo $117.6/ month$168 14× Lite usage All Pro benefits First access to the latest flagship models and features Dedicated resources during peak times Prices and plan benefits may change. Final details are shown on z.ai. Capabilities ## Stay on the Frontier [...] `git status --short`Failed $ ``` git status --short ``` ``` 2026-04-19 17:10:16.280 xcodebuild[60663:50240053] DVTFilePathFSEvents: Failed to start fs event stream.git: warning: confstr() failed w

## GLM-5.2 and Exclusive Benefits GLM-5.2 is now fully available in ZCode, with genuinely usable long-context capability suited to development tasks with longer scope, more files, and sustained reasoning. ZCode Agent continuously combines workspace state, tool results, and Git changes throughout task execution, turning model capability into stable code changes. Subscriber benefit #### Idle-time tasks run for free Subscribers can create idle-time tasks: queue up non-urgent work and ZCode completes it for free during periods of spare capacity, without consuming your plan quota. Rolling out gradually to subscribers. Execution costFree Plan quota consumed0 Learn about idle-time tasks New user offer #### 5-day free trial [...] With GLM-5.2's stable 1M context and long-horizon task capability, ZCode Agent can keep goals, files, terminal results, browser context, execution modes, and Git state in the same task. This allows complex development work to move from planning to implementat

ZCode Agent's deep integration turns these model capabilities into a practical coding experience: workspace understanding and file references feed the model high-quality context, execution modes control autonomy, and Goal Mode adds objective management, completion verification, and state recovery for long tasks — so the long-horizon abilities GLM-5.2 was trained for are fully released in the product. ## Workspace Entry ZCode Agent understands the ZCode workspace, task list, file references, model picker, execution modes, and Git branch state directly, making it a strong fit for the full development loop from understanding a request to editing code and checking changes before commit. [...] Stable 1M context: reliable long-task execution across a 1M-token context — the goal is not "fitting more tokens" but staying dependable along real coding-agent trajectories. Stronger long-horizon coding: approaches or surpasses several closed frontier models on long-task benchmarks, making it one