About billing tokens for AI chat
Assort tracks monthly AI usage with billing tokens: OpenAI API token counts adjusted by model-specific multipliers. Usage in the admin dashboard is shown in billing tokens.
Billing period
Calendar months in Japan Standard Time (JST), from the 1st 0:00 through the last moment of the month. This is separate from your Stripe billing cycle.
Monthly limits by plan
| Plan | Monthly billing token limit |
|---|---|
| Free | 300,000 |
| Starter | 1,000,000 |
| Team | 4,000,000 |
There is no per-user cap. Usage is counted per workspace.
Legacy Pro (existing subscribers)
Includes 5,000,000 billing tokens per month. Usage above the included amount is billed at ¥200 per 1,000,000 billing tokens (excluding tax).
When you reach the limit
On Free, Starter, and Team, AI chat is blocked for the rest of the month. Limits reset on the 1st (JST).
If you are approaching the limit, consider upgrading on the pricing page.
Model multipliers
Billing tokens = API token count × model multiplier. For example, with gpt-5.4-mini, 100 API tokens are recorded as 300 billing tokens.
| Model | Multiplier |
|---|---|
gpt-5.4-nano | 0.8x |
gpt-5.4-mini | 3x |
gpt-5.4 | 10x |
gpt-5.5 | 20x |
The default model is gpt-5.4-mini.
If an older model name remains in settings, usage is calculated as gpt-5.4-mini at runtime.
Threaded conversations
Continuing a thread in Slack sends prior messages to the API again, so each reply may consume more tokens.
View pricing plans