Skip to main content
Watch: I Built an AI App Without Leaving Databricks

Model APIs - 3 entries

Z.ai API and GLM Coding Plan limits

Z.ai's official docs, read on October 17, 2026, set limits for the Z.ai API and GLM Coding Plan. For GLM Coding Team Plan (per seat): GLM-5.3, GLM-5.3-Flash, plus MCP tools on Premium Seat, the docs list 35,000 5-Hour Credits and 155,000 Weekly Credits.

Last verified from docs.z.ai. What the docs leave out

Sources
2
Tables
3
Entries
3
Verified
Oct 17, 2026

Limits by model, endpoint and plan

Values are copied exactly as the docs show them, units included, in the docs' own order. A dash means the docs list no value for that row.

GLM Coding Team Plan (per seat): GLM-5.3, GLM-5.3-Flash, plus MCP tools

Premium SeatModel page: GLM-5.3-Flash

GLM Coding Team Plan (per seat): GLM-5.3, GLM-5.3-Flash, plus MCP tools limits
MetricPremium Seat
5-Hour Credits35,000
Weekly Credits155,000
ConcurrencyRelated to plan tier, dynamically adjusted; recommended for 2+ concurrent development projects
Estimated tokens/week, GLM-5.3, 95% cache hit749–1,497 million tokens/week
Estimated tokens/week, GLM-5.3-Flash, 95% cache hit2,267–4,533 million tokens/week
Estimated tokens/week, GLM-5.3, 96% cache hit768–1,536 million tokens/week
Estimated tokens/week, GLM-5.3-Flash, 96% cache hit2,326–4,652 million tokens/week
Estimated tokens/week, GLM-5.3, 98% cache hit810–1,619 million tokens/week
Estimated tokens/week, GLM-5.3-Flash, 98% cache hit2,454–4,908 million tokens/week
  • The estimated token table labels this column "Team Advanced" while the credit table calls it "Premium Seat". Premium Seat only: early access to new flagship models and "Priority access during peak hours".

GLM Coding Plan Legacy Plan V2 (individual, discontinued for new users July 30, 2026): GLM-5.3, GLM-5.3-Flash

Legacy Plan V2 Lite / Pro / MaxModel page: GLM-5.3-Flash

GLM Coding Plan Legacy Plan V2 (individual, discontinued for new users July 30, 2026): GLM-5.3, GLM-5.3-Flash limits
MetricLegacy Plan V2 Lite / Pro / Max
Lite 5-Hour LimitUp to approx. 80 prompts
Lite Weekly LimitUp to approx. 400 prompts
Pro 5-Hour LimitUp to approx. 400 prompts
Pro Weekly LimitUp to approx. 2,000 prompts
Max 5-Hour LimitUp to approx. 1,600 prompts
Max Weekly LimitUp to approx. 8,000 prompts
GLM-5.3 quota consumption rate1x during off-peak hours and 3x during peak hours
GLM-5.3-Flash quota consumption rate0.4x during off-peak hours and 1.2x during peak hours
  • "One prompt refers to one query. Each prompt is estimated to invoke the model 15–20 times." Monthly quota is "equivalent to approximately 15–30x the monthly subscription fee (weekly caps already factored in)". 5-hour limit dynamically refreshed (resets 5 hours after consumption); weekly limit resets every 7 days from subscription. Figures are estimates.

GLM Coding Team Plan subscribed before July 30, 2026 (legacy calculation, per seat)

Legacy Team Standard Seat / Premium Seat

GLM Coding Team Plan subscribed before July 30, 2026 (legacy calculation, per seat) limits
MetricLegacy Team Standard Seat / Premium Seat
Standard Seat quota per 5 hours, GLM-5.3Up to 60 M tokens
Standard Seat quota per 5 hours, GLM-5.3-FlashUp to 150 M tokens
Standard Seat weekly quota, GLM-5.3Up to 300 M tokens
Standard Seat weekly quota, GLM-5.3-FlashUp to 750 M tokens
Premium Seat quota per 5 hours, GLM-5.3Up to 160 M tokens
Premium Seat quota per 5 hours, GLM-5.3-FlashUp to 400 M tokens
Premium Seat weekly quota, GLM-5.3Up to 800 M tokens
Premium Seat weekly quota, GLM-5.3-FlashUp to 2,000 M tokens
  • "Maximum" refers to tokens consumable when the off-peak consumption multiplier applies. GLM-5.3 consumes 1x off-peak and 3x peak; GLM-5.3-Flash 0.4x off-peak and 1.2x peak. Peak hours Monday to Friday, 14:00–18:00 UTC+8.

How to raise your limits

  1. GLM Coding Team Plan (per seat): GLM-5.3, GLM-5.3-Flash, plus MCP tools

    On-demand usage overage (enabled by team administrator) with per-member spending limits; Premium Seat adds priority access during peak hours with fewer rate limits.

  2. GLM Coding Plan Legacy Plan V2 (individual, discontinued for new users July 30, 2026): GLM-5.3, GLM-5.3-Flash

    Legacy V2 holders can keep auto-renew and upgrade tier or billing term; switching immediately to the new credits-based plan is allowed only when upgrading to a higher tier.

  3. GLM Coding Team Plan subscribed before July 30, 2026 (legacy calculation, per seat)

    Cannot switch to the new credits-based plan before the current plan expires; subscribe to a current plan after expiry.

What the docs leave out

  • Standard pay-as-you-go API rate limits (RPM, TPM, concurrency per model and per usage tier) are not published in public docs: https://docs.z.ai/api-reference/rate-limit.md redirects to the logged-in console at https://z.ai/manage-apikey/rate-limits, and the help FAQ only points there. Only error codes 1302 (Rate limit reached for requests) and 1305 (temporarily overloaded) are public.
  • No documented process or criteria for how standard API usage tiers advance or how to request higher API rate limits.
  • GLM Coding Plan concurrency limits have no published numbers; docs only say they are dynamically adjusted with Max > Pro > Lite and raised off-peak.
  • Legacy Plan V1 usage limits are not listed in public docs (the notice only says they stay unchanged until the end of the billing cycle).
  • GLM-5.3-FlashX is not available on the Coding Plan; no plan quota published for it.

FAQ

What are the Z.ai API and GLM Coding Plan limits?

Z.ai's official docs, read on October 17, 2026, set limits for the Z.ai API and GLM Coding Plan. For GLM Coding Team Plan (per seat): GLM-5.3, GLM-5.3-Flash, plus MCP tools on Premium Seat, the docs list 35,000 5-Hour Credits and 155,000 Weekly Credits.

How do I raise my Z.ai API and GLM Coding Plan limits?

On-demand usage overage (enabled by team administrator) with per-member spending limits; Premium Seat adds priority access during peak hours with fewer rate limits. Legacy V2 holders can keep auto-renew and upgrade tier or billing term; switching immediately to the new credits-based plan is allowed only when upgrading to a higher tier. Cannot switch to the new credits-based plan before the current plan expires; subscribe to a current plan after expiry.

Where do these Z.ai API numbers come from?

Every value is copied verbatim from Z.ai's official documentation at docs.z.ai, read on October 17, 2026. This page lists all 3 entries in 3 tables, each linked to the page it came from.

Related reading

Sources

Last verified . Every value is copied verbatim from the official pages below, read on that date.

Newsletter

Know when limits change

One email per week. New model releases, pricing and limit changes, and what they mean for what you build.