StandardAnthropicLLMService¶
Description¶
A Controller Service that provides integration with Anthropic’s Claude AI models through their Messages API. Supports configurable parameters including model selection, response generation settings (temperature, top_p, top_k), token limits, and retry behavior.
Tags¶
ai, anthropic, api, claude, language model, llm, openflow
Properties¶
In the list below required Properties are shown with an asterisk (*). Other properties are considered optional. The table also indicates any default values, and whether a property supports the NiFi Expression Language.
| Display Name | API Name | Default Value | Allowable Values | Description |
|---|---|---|---|---|
| Anthropic API Key * | Anthropic API Key | The API Key for authenticating to Anthropic | ||
| Backoff Base Delay (ms) * | Backoff Base Delay (ms) | 1000 | The base delay in milliseconds for exponential backoff between retries | |
| Max Response Tokens * | Max Response Tokens | 1000 | The maximum number of tokens to generate in the response. | |
| Max Retries * | Max Retries | 3 | The maximum number of retry attempts for API calls | |
| Model Name * | Model Name | claude-3-5-sonnet-latest | The name of the Anthropic model | |
| Temperature | Temperature | The temperature to use for generating the response. | ||
| Top K | Top K | The top K value to use for generating the response. Only sample from the top K options for each subsequent token. Recommended for advanced use cases only. You usually only need to use temperature. | ||
| Top P | Top P | The top_p value for nucleus sampling. It controls the diversity of the generated responses. | ||
| User ID | User ID | The user id to set in the request metadata | ||
| Web Client Service * | Web Client Service | The Web Client Service to use for communicating with the LLM provider. |
State management¶
This component does not store state.
Restricted¶
This component is not restricted.
System Resource Considerations¶
This component does not specify system resource considerations.