Account & Billing
Billing Rules
Understand the platform's billing unit, the conversion between credits and tokens, and the deduction formula
Billing Unit: Credits
Credits (also referred to as "quota") are the platform's unified internal billing unit. After topping up, funds are converted to credits at the applicable rate. Each time you call an AI model, the system deducts the appropriate credits from your account based on the actual number of tokens consumed.
The conversion rate between platform credits and USD is shown on the top-up page.
Credits vs. Tokens
| Concept | Description |
|---|---|
| Token | The basic unit that large language models use to process text — roughly 1 English word ≈ 1 token; 1 Chinese character ≈ 1–2 tokens |
| Credits | The platform's billing unit, calculated as: token consumption × model rate × group multiplier |
Deduction Formula
Credits Consumed = (Input Tokens + Output Tokens × completion_multiplier) × model_rate × group_rate- Model rate: The unit price coefficient for a given model — higher-priced models have a higher rate
- Group rate: The discount coefficient for your group — higher-tier groups enjoy lower rates
- Completion multiplier: The price coefficient for output tokens (output is typically priced higher than input)
Real-Time Deduction
After each API call is completed, the system immediately calculates and deducts credits based on the actual token consumption. You can view deduction details in "Usage Logs".
For per-model rates, see the Model List & Capability Comparison page.
Last updated on