Z.ai has introduced ZCode as the dedicated cross-platform workspace built around the GLM-5.2 model, letting users orchestrate agents for planning, coding, review, and deployment while pulling in their own API keys.
Introducing ZCode, the official development environment for GLM-5.2
- GLM Coding Plan subscribers: now 1.5x usage quota in ZCode - BYOK supported: works with your existing subscriptions and APIs - Available on macOS, Windows, and Linux
Download now: http://zcode.z.ai/en
Anyone on the GLM Coding Plan now gets 1.5 times their normal allowance inside ZCode, though the exact length of this adjustment remains unspecified in the launch materials.
The app supports bring-your-own-key connections and messaging-app bot controls on macOS, Windows, and Linux, yet no head-to-head performance data against other harnesses has surfaced yet.
Many users praise ZCode's rapid launch and strong performance as the official IDE for GLM-5.2, while others criticize high pricing, bugs like reconnect loops, and a marketing focus over user needs.
No Digg Deeper questions have been answered for this story yet.
Follow @zcode_ai for the latest updates. See the full changelog: http://zcode.z.ai/en/changelog
@Zai_org 👋 That's me — Hi, I'm ZCode!
Made for GLM-5.2, ready when you are. Download → https://zcode-ai.com/ Say hi → https://x.com/zcode_ai
Introducing ZCode, the official development environment for GLM-5.2
- GLM Coding Plan subscribers: now 1.5x usage quota in ZCode - BYOK supported: works with your existing subscriptions and APIs - Available on macOS, Windows, and Linux
Download now: http://zcode.z.ai/en
@Zai_org @jietang Considering that GLM-5.2 is on par with Opus 4.6, and its pricing is much lower, it’s actually very competitive
@Zai_org It seens concerning that you are releasing closed source tools while advocating to open source :/
@Zai_org i’ve been using it it for like past week and the usage is very less the tokens we get for the free usage for like complex projects maybe just half an hour all gets exhausted so it’s of no use
@Zai_org Can we get a CLI please?
@Zai_org ZCode CLI when?
@lucasteske @Zai_org Brother can they have anything to themselves?
@Zai_org Does it have an in-app browser?
@Zai_org
@Zai_org Can you make it so it stops showing Chinese? I have the settings correct and the model displays Chinese. Can't read characters...
@Zai_org Full information here
@Zai_org Yo ZCode is awesome, downloaded it instantly and already loving the workflow.
@Zai_org I really like this idea, but unfortunately, I do not trust my data being processed in China. And on your page you are lacking private section. I can't find it.
@rezmeram @Zai_org Opencode is fine but not my CLI of choice.
Can’t put my finger on why, but I don’t use it.
I do reroute the plan to @FactoryAI using @openburnbar
@Zai_org Congratulations 🎉 Is it open source?
@BuffaloBarry716 @Zai_org The main reason is I like using multiple harnesses in CMUX.
I find it easier to orchestrate between models and harnesses when they all live in the same window.
@Zai_org @grok compare cost per task based on benchmarks like artificial analysis witht that 1.5x usage quota, compared to cursor and composer 2.5 and codex with gpt-5.5 for a $20 subscription
@Zai_org Damn, so you guys just cloned codex? lol
