> ## Documentation Index
> Fetch the complete documentation index at: https://docs.keyflow.space/llms.txt
> Use this file to discover all available pages before exploring further.

# AI Filter

### **Overview:**

The `AI Filter` block is designed to process input data by applying a specified natural language condition, facilitating effective data filtering and decision-making.

***

### Inputs & Outputs

| I/O    | Feature           | Type     | Simple Explanation                                                                                                                                                    |
| ------ | ----------------- | -------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| input  | `input`           | string   | The text to apply the filter on.                                                                                                                                      |
| input  | `condition`       | string   | The natural language condition that will dictate how the inputs are filtered.                                                                                         |
| input  | `model`           | selector | Select from a range of supported LLM models for carrying out inference based on provided inputs.                                                                      |
| input  | `cache_response`  | boolean  | When enabled, this feature saves AI calls by returning the same response if the inputs have not changed since the last processing instance.                           |
| output | `filtered_output` | string   | - This represents the filtered result of your data evaluation. If your input satisfies the defined condition, it appears here unchanged; otherwise, it will be empty. |

***

### Use Cases

Consider how this block can enhance various processes:

* **Filtering Customer Inquiries:** Efficiently identify and isolate customer inquiries that meet certain criteria expressed in everyday language, ensuring prompt responses.
* **Data Categorization:** Automatically sort and organize textual information into distinct categories based on predefined conditions—useful for managing large datasets or incoming messages.
* **Automated Decision Making:** Incorporate dynamic decision capabilities into automation workflows whereby textual content determines subsequent actions or outputs in a streamlined manner.

In essence, when there is a need to filter or make decisions based on textual data seamlessly, the `AI Filter` block emerges as an essential tool!