Zhipu GLM Coding Plan V3 uses a points system based on token consumption and offers personal and team tiers; official pages define current prices, point limits, and deduction rules.
Available | The V3 points-based plan is on sale. First-time subscribers and users without an active plan can purchase it directly; existing-plan rights remain protected under the official rules.
Available | The V3 points-based plan is on sale. First-time subscribers and users without an active plan can purchase it directly; existing-plan rights remain protected under the official rules.
Available | The V3 points-based plan is on sale. First-time subscribers and users without an active plan can purchase it directly; existing-plan rights remain protected under the official rules.
Credit (5h/7d/1m)
2,000 / 10,000 / -
12,000 / 60,000 / -
28,000 / 140,000 / -
Agent concurrency
-
-
-
Supported models
Harness tools
Web Reader MCP, Web Search MCP, Open-source Repository MCP, Visual Understanding MCP
Open-source Repository MCP, Web Search MCP, Web Reader MCP, Visual Understanding MCP
Open-source Repository MCP, Web Reader MCP, Web Search MCP, Visual Understanding MCP
Notes
GLM Coding Plan V3 Personal Lite costs CNY 118/month with 2,000 points per rolling 5 hours and 10,000 points per week.
Weekly token reference range: Official reference (95% cache, weekly credits): GLM-5.3 about 48–97M tokens/week; GLM-5.3-Flash about 146–292M tokens/week.
OpenClaw and GLM in Excel: OpenClaw is supported with secondary scheduling and best-effort delivery; Coding Agent tasks have preemption priority. Under high load, OpenClaw may queue or rate-limit. Plans also include GLM in Excel (Beta).
Dual quota windows: Rolling 5-hour and weekly limits both apply. Five-hour points recover dynamically five hours after consumption, while weekly points refresh every seven days from purchase.
Peak and off-peak deductions: Off-peak calls consume 50% of the base points. Peak hours are Monday through Friday, 14:00–18:00 (UTC+8).
Quota exhaustion behavior: When Coding Plan points are exhausted, other resource packages or account balance are not consumed automatically.
GLM Coding Plan V3 Personal Pro costs CNY 538/month with 12,000 points per rolling 5 hours and 60,000 points per week.
Weekly token reference range: Official reference (95% cache): GLM-5.3 about 290–580M tokens/week; GLM-5.3-Flash about 877–1755M tokens/week.
OpenClaw and GLM in Excel: OpenClaw is supported with secondary scheduling and best-effort delivery; Coding Agent tasks have preemption priority. Under high load, OpenClaw may queue or rate-limit. Plans also include GLM in Excel (Beta).
Dual quota windows: Rolling 5-hour and weekly limits both apply. Five-hour points recover dynamically five hours after consumption, while weekly points refresh every seven days from purchase.
Peak and off-peak deductions: Off-peak calls consume 50% of the base points. Peak hours are Monday through Friday, 14:00–18:00 (UTC+8).
Quota exhaustion behavior: When Coding Plan points are exhausted, other resource packages or account balance are not consumed automatically.
GLM Coding Plan V3 Personal Max costs CNY 1,078/month with 28,000 points per rolling 5 hours and 140,000 points per week.
Weekly token reference range: Official reference (95% cache): GLM-5.3 about 676–1352M tokens/week; GLM-5.3-Flash about 2047–4095M tokens/week.
OpenClaw and GLM in Excel: OpenClaw is supported with secondary scheduling and best-effort delivery; Coding Agent tasks have preemption priority. Under high load, OpenClaw may queue or rate-limit. Plans also include GLM in Excel (Beta).
Dual quota windows: Rolling 5-hour and weekly limits both apply. Five-hour points recover dynamically five hours after consumption, while weekly points refresh every seven days from purchase.
Peak and off-peak deductions: Off-peak calls consume 50% of the base points. Peak hours are Monday through Friday, 14:00–18:00 (UTC+8).
Quota exhaustion behavior: When Coding Plan points are exhausted, other resource packages or account balance are not consumed automatically.
Available | The new points-based team plan is on sale. Existing team subscriptions retain their current-cycle rights and can move to the new online plan after expiry.
Available | The new points-based team plan is on sale. Existing team subscriptions retain their current-cycle rights and can move to the new online plan after expiry.
Credit (5h/7d/1m)
15,000 / 66,000 / -
35,000 / 155,000 / -
Agent concurrency
-
-
Supported models
Harness tools
Visual Understanding MCP, Open-source Repository MCP, Web Search MCP, Web Reader MCP
Open-source Repository MCP, Web Search MCP, Web Reader MCP, Visual Understanding MCP
Notes
GLM Coding Plan V3 Team Standard costs CNY 598 per seat/month with 15,000 points per rolling 5 hours and 66,000 points per week.
Suggested project usage: The official project-count guidance is a usage recommendation, not an Agent concurrency limit.
Weekly token reference range: Official reference (95% cache): GLM-5.3 about 319–638M tokens/week; GLM-5.3-Flash about 965–1930M tokens/week.
OpenClaw and GLM in Excel: OpenClaw is supported with secondary scheduling and best-effort delivery; Coding Agent tasks have preemption priority. Under high load, OpenClaw may queue or rate-limit. Plans also include GLM in Excel (Beta).
Dual quota windows: Rolling 5-hour and weekly limits both apply. Five-hour points recover dynamically five hours after consumption, while weekly points refresh every seven days from purchase.
Peak and off-peak deductions: Off-peak calls consume 50% of the base points. Peak hours are Monday through Friday, 14:00–18:00 (UTC+8).
Quota exhaustion behavior: When Coding Plan points are exhausted, other resource packages or account balance are not consumed automatically.
GLM Coding Plan V3 Team Advanced costs CNY 1,198 per seat/month with 35,000 points per rolling 5 hours and 155,000 points per week.
Suggested project usage: The official project-count guidance is a usage recommendation, not an Agent concurrency limit.
Weekly token reference range: Official reference (95% cache): GLM-5.3 about 749–1497M tokens/week; GLM-5.3-Flash about 2267–4533M tokens/week.
OpenClaw and GLM in Excel: OpenClaw is supported with secondary scheduling and best-effort delivery; Coding Agent tasks have preemption priority. Under high load, OpenClaw may queue or rate-limit. Plans also include GLM in Excel (Beta).
Dual quota windows: Rolling 5-hour and weekly limits both apply. Five-hour points recover dynamically five hours after consumption, while weekly points refresh every seven days from purchase.
Peak and off-peak deductions: Off-peak calls consume 50% of the base points. Peak hours are Monday through Friday, 14:00–18:00 (UTC+8).
Quota exhaustion behavior: When Coding Plan points are exhausted, other resource packages or account balance are not consumed automatically.
Purchase and data notes
This page is for reference only. Official provider pricing, benefits, terms, and service agreements take precedence.
Coding Plans, Token Plans, and resource packages can have different permitted uses. Review the provider's rules before buying.
Request counts, token usage, and stability vary by model, task, and tool. Validate your workload with a low-cost plan or free quota first.
Pricing, supported models, limits, and availability can change. Use report an issue to help us verify and update the data.
This page is for reference only. Official provider pricing, benefits, terms, and service agreements take precedence.
Coding Plans, Token Plans, and resource packages can have different permitted uses. Review the provider's rules before buying.
Request counts, token usage, and stability vary by model, task, and tool. Validate your workload with a low-cost plan or free quota first.
Pricing, supported models, limits, and availability can change. Use report an issue to help us verify and update the data.
