Every known z.ai bonus window, multiplier and limited-time discount — computed live in your browser with ticking countdowns, timezone-aware timelines and a Prometheus-compatible metrics endpoint. No stale screenshots, no guessing when to fire the heavy agent runs.
your local time
—
schedule time
— Asia/Singapore · UTC+8
live status · ai-synced hourly
Status badges flip automatically as windows open and close. Countdowns tick every second; schedules are evaluated against Asia/Singapore time exactly as the official notices define them — and an AI bot re-verifies them against docs.z.ai every hour.
z.ai
Time-variable rates apply during peak hours. Code off-peak and the same credits last twice as long.
z.ai
Nightly bonus window for GLM-5.3-Flash: unlimited usage in ZCode and doubled quota in other supported agents.
z.ai
Limited-time 50% discount on GLM-5.3-Flash API pricing — the list price is slashed across input, cached input and output tokens.
planner
The same 24 hours shown twice: once in schedule time (SGT), once in your local clock. Line up the bright segments and you never accidentally pay peak rates again.
Bars start at midnight in each timezone. The peak surcharge only runs Mon–Fri; the flash campaign paints the whole night. Wherever you are, the night bar is when the stack is deepest.
planner · utc+8
z.ai is built in Beijing — during China's public holidays (Golden Week, Spring Festival and friends) support replies, pricing updates and non-critical releases slow down across China-based companies. Plan heavy work and deadline-sensitive runs around these windows; the API itself stays up.
z.ai is built in Beijing, and its teams observe China's public holidays — during Golden Week, Spring Festival and the shorter festivals, support replies and non-critical releases slow down across China-based companies while the API itself stays on-call.
China has 13 statutory holiday days a year, extended into week-long breaks by swapping weekends — the swapped-in Saturday/Sundays are working days with normal operations. 2026 dates follow the official State Council notice; later years are the expected pattern until published (usually each November). API infrastructure stays on-call through every holiday — this affects support, pricing updates and release pace far more than uptime.
pay as you go
Prices per 1M tokens in USD. Limited-time API promos are folded in the moment the official pricing page publishes them — the hourly AI sync keeps this table honest around the clock.
Flagship coding model
per 1M tokens · cached input storage free for a limited time
Fast · 50% off for a limited time
per 1M tokens · cached input storage free for a limited time
| model | input / cached input per 1M | output per 1M |
|---|---|---|
| GLM-5.2 | $1.4 / $0.26 | $4.4 |
| GLM-5.1 | $1.4 / $0.26 | $4.4 |
| GLM-5 | $1.0 / $0.20 | $3.2 |
| GLM-4.7 | $0.6 / $0.11 | $2.2 |
| GLM-4.5-Air | $0.2 / $0.03 | $1.1 |
| GLM-4.7-FlashX | $0.07 / $0.01 | $0.4 |
| GLM-4.6V (vision) | $0.3 / $0.05 | $0.9 |
| GLM-4.7-Flash | Free | Free |
Full price list at docs.z.ai/guides/overview/pricing ↗
subscriptions
Credits-based plans for Claude Code, Cline, OpenCode, ZCode and more. Off-peak usage bills at 50% of the standard credit rate — the same window the radar tracks above.
Daily driving for solo developers — quick fixes, codebase Q&A, completion-heavy work.
5-hour credits
2,000
weekly credits
10,000
High-frequency feature work on real repositories — 6× the Lite allowance in both windows.
5-hour credits
12,000
weekly credits
60,000
All-day agentic coding on large codebases, long refactors and parallel agent fleets.
5-hour credits
28,000
weekly credits
140,000
Credits refresh dynamically: the 5-hour bucket resets 5 hours after consumption, the weekly bucket resets every 7 days. Credit usage = (input × input multiplier + cached input × cached multiplier + output × output multiplier) ÷ 10,000 — and everything billed off-peak costs half. Source: GLM Coding Plan overview ↗
invite token · first order only
Subscribe through this invite token and the discount is applied instantly at checkout — no manual activation. Valid for new users and existing accounts that have never had a paid subscription, on the first GLM Coding order only.
toolkit · by the same builder
Field-tested utilities and deep-dive write-ups from Rommark.Dev — the GitHub auto-push protocol that keeps agent work safe, the self-healing deploy kit that keeps this very portal online, and hands-on reviews of agent memory and multi-agent orchestration tooling.
Step-by-step tutorial on generating GitHub PAT tokens, setting automated workspace & chat-history push rules in web and local IDEs, and restoring projects across sessions — how an AI agent's work survives the sandbox.
claw.rommark.devgithub-auto-pushSelf-healing deploy kit for z.ai fullstack hosting — repairs missing build artifacts, CAExited boot errors, pack/rebuild races and the “Sorry, there was a problem deploying the code” doom loop. Battle-tested on this very portal.
github.comzai-hosting-reviverThe radar's own scanner suite, five online scanners with live progress + ETA: Security Audit (headers, CORS, exposure probes, leaked secrets), SEO/GEO/Performance (AI-crawler policy, llms.txt, TTFB, payload), a QA reliability audit, an Auto QA Auditor that drives the app in a real browser end-to-end (asks for logins before the run, pauses to ask questions mid-run, runs axe a11y, 375px responsive, perf timings) and a heuristic Code Reviewer. Each accepts a website/IP or a GitHub repo (private repos via a token kept in memory only), re-verifies findings before reporting, and ends with a copy-paste fix prompt for your dev agent.
this appzscannerThe radar's own GEO-first search engine: it searches the live web the way generative engines read it. Instead of SEO, every result is scored for citability, structure, freshness and entity clarity — and an AI answer preview shows which sources actually get cited inside AI answers.
this appzsearchThe radar's built-in arcade: thirty playable 8-bit classics — snake, blocks, breakout, pong, invaders, asteroids, hopper, muncher, kong, lander, missile, racer, cycle, bomber, cubes, flap, runner, merge, mines, simon, memory, whack, stacker, xo, ships, slots, pinball, hangman, punch and maze — rendered on pixel canvases in your browser. Keyboard on desktop, touch pads on the phone, best scores stored locally and synced with nothing.
this appzfunPersistent session memory and MCP vector context enhancer for Claude Code, Cursor, Windsurf, Cline and chat.z.ai — 33 tools, GitHub storage sync and token compression, reviewed in depth.
claw.rommark.devzaimemGVS5H ledger orchestration reviewed: disk-backed multi-agent collaboration with adversarial test specs — the technique that matched Claude Fable 5 on LiveCodeBench-Hard.
claw.rommark.devzcode-smart-skillThe GVS5H shared-filesystem-ledger technique shipped as a drop-in /smart command for ZCode with auto-triggering — five open-weight Qwen3.8-27B instances matching Claude Fable 5.
claw.rommark.devzcode-smart-mode