Usage limits
Updated
The Orchestrate conversational interface processes text in small chunks called tokens, roughly one token per word, though punctuation and spaces also count. Orchestrate tracks token usage to manage how much information you can send to in the chat and how much it can send back.
- Input tokens: Units to measure the amounts of text in your prompt, any data you provide, and the agent's instructions.
- Output tokens: Units to measure the amount of text in the agent's response.
The following table lists token limits that apply to each prompt and response:
Per-request token limits
| Dimension | Limit | Description |
|---|---|---|
| Input tokens | ~ 200,000 tokens | Total capacity of the user request including goal, clarifications, mode specifications, and Meta Hub glossaries |
| Response tokens | ~ 16,000 tokens | Maximum length of a single Orchestrate reply |
Per-account rate quotas
Account limitations on tokens refresh after the rolling time window.
| Window | Cumulative Token Count | Notes |
|---|---|---|
| Hourly | 1,000,000 tokens | Resets every hour (rolling hourly window). Each design session is approximately 200k tokens |
| Daily | 5,000,000 tokens | Resets daily at 00:00:00 UTC. |
| Monthly | 40,000,000 tokens | Resets on the 1st of each month at 00:00:00 UTC. |
How to check token usage
Select the information icon in the conversational panel to display token usage statistics.
