Z.ai subscriptions
GLM Coding Plan subscription pricing
Compare 3 GLM Coding Plan plans, included models, usage notes, and when it makes more sense to use API token pricing instead.
Data updated .
GLM Coding Plan limits and pricing at a glance
The GLM Coding Plan is Z.ai's flat-fee subscription for using GLM-5.2, GLM-5-Turbo, GLM-4.7, and GLM-4.5-air inside coding tools like Claude Code, Cline, Roo Code, OpenClaw, and 20+ other supported clients. Instead of paying per token through the Z.ai API, you pay one monthly fee and get a fixed quota of prompts.
- Lite - $18/mo ($12.60/mo promo, $151.20/yr): up to about 80 prompts per 5 hours and about 400 prompts per week, plus 100 MCP web-search/reader calls per month.
- Pro - $72/mo ($50.40/mo promo, $604.80/yr): up to about 400 prompts per 5 hours and about 2,000 prompts per week, plus 1,000 MCP calls per month. That is about 5x Lite usage.
- Max - $160/mo ($112/mo promo, $1,344/yr): up to about 1,600 prompts per 5 hours and about 8,000 prompts per week, plus 4,000 MCP calls per month. That is about 20x Lite usage.
All three tiers include the same model lineup. The differences are quota size, MCP allowance, and priority during peak demand.
How GLM Coding Plan quota actually counts
Z.ai defines one prompt as one user query, but each prompt may invoke the model 15-20 times behind the scenes. The 5-hour limit is dynamically refreshed and resets 5 hours after quota consumption. The weekly limit starts when the subscription is activated and resets every 7 days.
GLM-5.2 and GLM-5-Turbo use more quota than routine models: 3x during peak hours and 2x off-peak. Z.ai lists peak hours as 14:00-18:00 UTC+8. As a limited-time benefit through the end of September, GLM-5.2 and GLM-5-Turbo only consume 1x quota off-peak. For routine edits, GLM-4.7 can preserve quota.
These are estimates, not guaranteed message counts. Actual usage can shrink when a repository is large, the task is complex, or auto-accept causes the agent to keep working through more model calls.
GLM Coding Plan vs the Z.ai API
If you fire off occasional calls from a script, the Z.ai API (pay per million tokens) is cheaper. If you live inside an agent IDE for hours a day, the GLM Coding Plan caps your bill and almost always wins on $/token. Use the subscription vs API calculator to model your personal break-even point based on daily prompt count and average tokens per call.
How does GLM Coding Plan compare to Claude Max and ChatGPT Pro?
At $160/mo, GLM Coding Plan Max sits below ChatGPT Pro ($200/mo) and Claude Max's higher tiers. For pure coding workloads where you are happy with GLM-5.2 instead of Claude or GPT-5, it is currently the cheapest high-quota plan we track for agentic IDE use. For general chat, research, multimodal, or non-coding work, Anthropic and OpenAI subscriptions still cover more ground.
Cheapest paid plan
GLM Coding Lite
$18/mo
Highest self-serve plan
GLM Coding Max
$168/mo
Cheapest related API model
GLM-5.2
$1.4/1M input tokens
| Plan | Price | Included models | Usage notes |
|---|---|---|---|
| GLM Coding Lite lite | $18/mo | glm-5.2, glm-5-turbo, glm-4.7, glm-4.5-air | GLM Coding Plan is limited to officially supported coding tools. Lite includes up to about 80 prompts per 5 hours and 400 prompts per week. GLM-5.2 and GLM-5-Turbo consume 3x quota during peak hours and 2x off-peak, with a limited-time 1x off-peak benefit through September. |
| GLM Coding Pro pro | $80/mo | glm-5.2, glm-5-turbo, glm-4.7, glm-4.5-air | GLM Coding Plan is limited to officially supported coding tools. Pro includes up to about 400 prompts per 5 hours and 2,000 prompts per week, about 5x Lite usage. |
| GLM Coding Max max | $168/mo | glm-5.2, glm-5-turbo, glm-4.7, glm-4.5-air | GLM Coding Plan is limited to officially supported coding tools. Max includes up to about 1,600 prompts per 5 hours and 8,000 prompts per week, about 20x Lite usage. |
Break-even calculator
Compare GLM Coding Plan plans vs API usage
Enter your daily message count and token volume to see whether a monthly subscription or raw API usage is cheaper.
API pricing
View Z.ai API token pricing
Compare per-million-token input, output, cached input, and batch prices for programmatic workloads.
Frequently asked questions
Does AI Pricing Guru have a GLM Coding Plan referral discount?
Yes. Our Z.ai invite link gives 10% off your first GLM Coding Plan subscription. Discounts are subject to Z.ai checkout rules, so we do not claim the invite discount stacks with the separate 30% introductory promo.
What are the GLM Coding Plan limits?
GLM Coding Plan Lite ($18/mo) allows roughly 80 prompts per 5 hours and 400 prompts per week, with 100 MCP web-search/reader calls per month. Pro ($72/mo) raises that to about 400 prompts per 5 hours, 2,000 per week, and 1,000 MCP calls. Max ($160/mo) gets about 1,600 prompts per 5 hours, 8,000 per week, and 4,000 MCP calls. Z.ai says each prompt may invoke the model 15-20 times, so real usage varies by project complexity, repository size, and auto-accept behavior.
How much does the GLM Coding Plan cost?
GLM Coding Plan Lite is $18/month, Pro is $72/month, and Max is $160/month. Z.ai also offers an introductory 30% discount that drops the prices to $12.60, $50.40, and $112 per month respectively, plus yearly billing at $151.20, $604.80, and $1,344 per year for the three tiers.
Which models are included in the GLM Coding Plan?
All three GLM Coding Plan tiers (Lite, Pro, Max) include the same model lineup: GLM-5.2 (the current flagship), GLM-5-Turbo, GLM-4.7, and GLM-4.5-air. GLM-5.2 and GLM-5-Turbo consume 3x quota during the 14:00-18:00 UTC+8 peak window and 2x off-peak, with a limited-time 1x off-peak benefit through the end of September.
Which coding tools support the GLM Coding Plan?
The GLM Coding Plan ships an OpenAI-compatible endpoint that drops into 20+ supported coding assistants, including Claude Code, Cline, Roo Code, OpenClaw, and other IDE integrations. The plan is restricted to officially supported coding tools — it is not a general Z.ai API key replacement.
Is the GLM Coding Plan cheaper than the Z.ai API?
It depends on usage. The Z.ai API is pay-per-token, so light or bursty usage is cheaper through the API. The GLM Coding Plan caps your bill at $18, $72, or $160 per month, so heavy daily coding sessions are almost always cheaper on a subscription. The break-even is roughly when your daily token spend on the API exceeds the monthly cap divided by 30 — use the subscription vs API calculator to model your specific workload.
How does GLM Coding Plan Max compare to Claude Max and ChatGPT Pro?
GLM Coding Plan Max is $160/month, slightly cheaper than ChatGPT Pro at $200/month and Claude Max 20x. For pure coding workloads where GLM-5.2 quality is acceptable, GLM Coding Plan Max is currently the cheapest high-quota agentic IDE plan we track. Claude Max and ChatGPT Pro cover broader use cases (research, multimodal, general chat) and bring Claude Sonnet/Opus and GPT-5 respectively.
Can I use the GLM Coding Plan for production API workloads?
No. The GLM Coding Plan is restricted to officially supported coding tools and IDE assistants. For programmatic, production, or customer-facing workloads, use the Z.ai API, which is metered per million tokens and has no tool restriction.