diff --git a/README.md b/README.md index fc27e152..de68e071 100644 --- a/README.md +++ b/README.md @@ -235,7 +235,7 @@ Dynamic settings can include the following parameters: | keys. | `project`: Project name is assigned to this key. **Required**
`role`: a role to be assigned to the key. **Note:** a key is invalid if `role` and `roles` are missed.
`roles`: a list of roles to be assigned to the key. **Note:** a key is invalid if `role` and `roles` are missed.
`secured`: the flag indicates if the key is secured. If it's set to `true` user request and deployment response won't be saved to the prompt log storage. | | roles | API key or user roles. Each role may have limits to be associated with applications, models, assistants or addons. Refer to [API Keys](https://github.com/epam/ai-dial/blob/main/docs/Roles%20and%20Access%20Control/3.API%20Keys.md) to learn more. | | roles. | `limits`: Limits for models, applications, or assistants. **Note:** it is necessary to define this for a role. | -| roles..limits | `minute`: Total tokens per minute limit sent to the model, managed via floating window approach for well-distributed rate limiting. If it's not set the default value is unlimited
`day`: Total tokens per day limit sent to the model, managed via floating window approach for balanced rate limiting. **Note**: you can skip these parameters to apply their default value - unlimited. | +| roles..limits | `minute`: Total tokens per minute limit sent to the model, managed via floating window approach for well-distributed rate limiting. If it's not set the default value is unlimited
`day`: Total tokens per day limit sent to the model, managed via floating window approach for balanced rate limiting.
`week`: Total tokens per week limit sent to the model, managed via floating window approach for balanced rate limiting.
`month`: Total tokens per month limit sent to the model, managed via floating window approach for balanced rate limiting.
**Note**: you can skip these parameters to apply their default value - unlimited. | | retriableErrorCodes | List of retriable error codes for handling outages at LLM providers. | ## License