Skip to main content

​
Pricing

Pricing is plan-based: every plan includes a request allowance. On-demand pricing applies only to usage beyond your plan’s included allowance.

​
How much of your allowance a call uses

Not every model costs the same slice of your plan. Each model has a quota multiplier: Auto and flash models cost 1.0 (the floor); GLM-5.2 costs 2.0; mid-tier models sit between; premium models cost more. A GLM-5.3 request whose prompt exceeds 262,144 tokens costs 4.0. Image models use a fixed multiplier per image — see POST /v1/images/generations. See Model Request Ratios for the full table, including how many requests each plan buys per model. Two ways to read it at runtime:
  • GET /v1/models reports quota_multiplier on every model, plus quota_context_surcharge on models whose cost scales with prompt size.
  • Every response carries X-Quota-Cost (what that call used) and X-Quota-Remaining (roughly what is left in the window).
This is separate from the on-demand token rates below, which only apply once your included allowance is gone. For on-demand usage, every model has explicit input and output rates in USD per million tokens. Rates are shown on:
  • The public Models page
  • Your dashboard usage history (per-request cost)
On-demand costs are calculated as:
Separate input and output rates.