602 B
602 B
InferenceAnthropicServiceSettings [interface-InferenceAnthropicServiceSettings]
| Name | Type | Description |
|---|---|---|
api_key |
string | A valid API key for the Anthropic API. |
model_id |
string | The name of the model to use for the inference task. Refer to the Anthropic documentation for the list of supported models. |
rate_limit |
InferenceRateLimitSetting | This setting helps to minimize the number of rate limit errors returned from Anthropic. By default, the anthropic service sets the number of requests allowed per minute to 50. |