Rate Limits
This guide describes the default API rate limits on a per workspace basis.
We establish rate limits in Jasper's API to safeguard against abuse and to ensure optimal performance and stability for all of our customers using the API.
Rate limits are on a per-workspace basis, and the default values are as follows:
Request Method | Endpoint | Rate limit |
---|---|---|
POST | /commands | 105 RPM (6,300 per hour) |
POST | /templates/:id/run | 105 RPM (6,300 per hour) |
POST | /keep-writing | 105 RPM (6,300 per hour) |
GET | /templates | 200 RPM (12,000 per hour) |
GET | /templates/:id | 200 RPM (12,000 per hour) |
If you expect to scale past these rate limits, please reach out to your Customer Success Manager so that the API team can coordinate with your engineering team, and increase your workspace's rate limits if needed.
AI Engine Fallbacks
In order to circumvent third-party model provider rate limits, downtime, or performance degradation, the Jasper's AI Engine configures a priority-order of fallbacks on a per endpoint, per-use case (e.g template) basis. This feature is intended to ensure integration is not impacted by downstream LLM provider issues or rate limits.
Updated about 1 year ago