Project stages and rate limits
On TaskingAI, projects may have different stages, including development, production. Each stage has different functionalities and rate limits.
Project stages
Currently, TaskingAI supports two project stages: Development and Production.
Development stage is for testing and developing purposes, while Production stage is for deploying and serving your AI Applications.
Production stage projects are applied with L1 rate limits, while Development stage projects are applied with L0 rate limits. For detailed information on the rate limits, please refer to the following section.
Rate Limits
For each project, we limit the number of requests per minute (RPM) made by API Keys. All API Keys under the same project contributes. The rate limits are divided into two levels: L0 and L1. L1 level rate limits are usually applied to projects at production stage, while L0 level rate limits are applied to projects at development stage.
Below are the rate limits for each API route, grouped by modules:
- Retrieval
- Inference
- Assistant
- Tool
- File
Route | Method | L0 Rate Limit (RPM) | L1 Rate Limit (RPM) |
---|---|---|---|
/collections | GET | 20 | 2000 |
POST | 20 | 500 | |
/collections/{collection_id} | GET | 60 | 6000 |
POST | 20 | 500 | |
DELETE | 20 | 500 | |
/collections/{collection_id}/records | GET | 20 | 2000 |
POST | 20 | 500 | |
/collections/{collection_id}/records/{record_id} | GET | 60 | 6000 |
POST | 20 | 500 | |
DELETE | 20 | 500 | |
/collections/{collection_id}/chunks | GET | 60 | 6000 |
POST | 30 | 500 | |
/collections/{collection_id}/chunks/{chunk_id} | GET | 60 | 6000 |
POST | 30 | 500 | |
DELETE | 30 | 500 | |
/collections/{collection_id}/chunks/query | POST | 20 | 500 |
Route | Method | L0 Rate Limit (RPM) | L1 Rate Limit (RPM) |
---|---|---|---|
/inference/text_embedding | POST | 60 | 5000 |
/inference/chat_completion | POST | 60 | 5000 |
Route | Method | L0 Rate Limit (RPM) | L1 Rate Limit (RPM) |
---|---|---|---|
/assistants | GET | 20 | 2000 |
POST | 20 | 500 | |
/assistants/{assistant_id} | GET | 60 | 6000 |
POST | 20 | 500 | |
DELETE | 20 | 500 | |
/assistants/{assistant_id}/chats | GET | 20 | 2000 |
POST | 20 | 500 | |
/assistants/{assistant_id}/chats/{chat_id} | GET | 60 | 6000 |
POST | 20 | 500 | |
DELETE | 20 | 500 | |
/assistants/{assistant_id}/chats/{chat_id}/messages | GET | 20 | 2000 |
POST | 20 | 500 | |
/assistants/{assistant_id}/chats/{chat_id}/messages/{message_id} | GET | 60 | 6000 |
POST | 20 | 500 | |
/assistants/{assistant_id}/chats/{chat_id}/generate | POST | 20 | 500 |
Route | Method | L0 Rate Limit (RPM) | L1 Rate Limit (RPM) |
---|---|---|---|
/actions | GET | 20 | 2000 |
/actions/{action_id} | GET | 60 | 6000 |
POST | 20 | 500 | |
DELETE | 20 | 500 | |
/actions/bulk_create | POST | 20 | 500 |
/actions/{action_id}/run | POST | 20 | 500 |
Route | Method | L0 Rate Limit (RPM) | L1 Rate Limit (RPM) |
---|---|---|---|
/files | POST | 3 | 100 |