> For the complete documentation index, see [llms.txt](https://docs-conversational-ai.syntphony.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs-conversational-ai.syntphony.com/configurations/value-ranges.md).

# Value Ranges

Use **Value Ranges** to fine-tune project behavior through predefined range-based configurations.

A Value Range controls a measurable limit used by Syntphony CAI during execution, such as how long the project waits for a generative AI response or how many tokens a supported model can use.

**Value Ranges are different from Parameters and Variables**, so consider that:

* **Value Ranges** control predefined project behavior through measurable values;
* **Parameters** store static reusable values;&#x20;
* **Variables** store values that can change during conversation execution.

{% hint style="info" %}
Use Value Ranges when you need to adjust operational limits that affect how the project behaves at runtime.
{% endhint %}

## Where to find Value Ranges?

Open the project and go to:

**Settings > Value ranges**

The Value Ranges page displays configuration groups, that includes:

* **Request timeout for generative AI**
* **Tokens limit for generative AI**

Each group contains cards for the modules that support that configuration:

<figure><img src="https://4008706377-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fn6zS4HeuuVpRHZEvDiFU%2Fuploads%2FVkOF9sGl7rTM0gH6Jfjk%2Fimage.png?alt=media&amp;token=9dbfddfe-99fd-4215-9d2d-65fd92473086" alt=""><figcaption><p>Value Ranges page </p></figcaption></figure>

### Request timeout for generative AI

**Request timeout for generative AI** defines how long Syntphony CAI waits for a response from a supported generative AI capability. The value is configured in seconds.

Use this setting when you need to control how long the project should wait before applying the documented timeout behavior.

### What each feature does

<table data-search="false"><thead><tr><th>Card</th><th align="right">Unit</th><th width="250">Tooltip description</th></tr></thead><tbody><tr><td>Zero-Shot LLM Model <strong>(request timeout for generative AI)</strong></td><td align="right">seconds</td><td>Set a value in seconds to wait for a response from the provider. If the request exceeds the time set, the system will trigger the Not Expected flow.</td></tr><tr><td>Prompt cell</td><td align="right">seconds</td><td>Set a value in seconds to wait for a response from the provider. If the request exceeds the time set, the flow will be halted.</td></tr><tr><td>Knowledge</td><td align="right">seconds</td><td>Set the timeout, in seconds, to wait for a response from the LLM. If the timeout is reached, the system triggers the Not Expected flow.</td></tr><tr><td>Rephrase Answer</td><td align="right">seconds</td><td>Set a value in seconds to wait for a response from the provider. If the request exceeds the time set, the system will deliver a static answer.</td></tr><tr><td>Multilingual Agent</td><td align="right">seconds</td><td><p>Set the timeout, in seconds, to wait for a response from the translation service:</p><p></p><ul><li><strong>NLU-first configuration:</strong> If the timeout is reached while interpreting the user's input, the system returns the <strong>Not Expected</strong> flow.</li><li><strong>Agentic-first and Composite configurations:</strong> If the timeout is reached while interpreting the user's input, the system returns either a Virtual Agent error response or a generic error.</li></ul><p></p><p><strong>Output translation:</strong> If the timeout is reached while translating the Virtual Agent's output, the Virtual Agent returns the response in the primary language.</p></td></tr><tr><td>MCP Server</td><td align="right">seconds</td><td>Set a value in seconds to wait for a response from the MCP Server. If the request exceeds the time set, the system will deliver the configured error answer or a generic error.</td></tr><tr><td>Zero-Shot LLM Model <strong>(Tokens limit for generative AI)</strong></td><td align="right">tokens</td><td>Adjust the value to determine the agent’s token consumption capacity.</td></tr></tbody></table>

### Tokens limit for generative AI

**Tokens limit for generative AI** defines the maximum number of tokens available for a supported generative AI configuration.

In the interface, the documented card under this group is Zero- Shot.&#x20;

Zero-Shot models are affected by token limits. When the configured token limit is reached, Syntphony CAI disables the functions for importing or creating new intents.

<div><figure><img src="https://4008706377-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fn6zS4HeuuVpRHZEvDiFU%2Fuploads%2FimahHkq9KWeYwhKLpoJz%2Fimage.png?alt=media&amp;token=95c8e83b-6278-4b82-a48f-3360aed43e77" alt=""><figcaption><p>Zero-Shot card </p></figcaption></figure> <figure><img src="https://4008706377-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fn6zS4HeuuVpRHZEvDiFU%2Fuploads%2FdCfEgeQrjyIf6m5SnQuk%2Fimage.png?alt=media&amp;token=f53de8d8-2e24-4eea-b186-de8a73ae56f5" alt=""><figcaption><p>Zero-Shot tooltip description for token consumption</p></figcaption></figure></div>

## What happens after changing a Value Range

After the value is changed, Syntphony CAI uses the new configuration when the affected module runs.

For instance:

* A shorter timeout may make the project stop waiting sooner;
* A longer timeout may give the module more time to respond;
* A lower token limit may reduce the available token capacity for the supported model;
* Reaching the Zero-Shot token limit disables importing or creating new intents.

{% hint style="info" %}
Value Ranges affect project behavior during execution. Review and test changes before applying them to projects used by real users.
{% endhint %}

## Troubleshooting

| Issue type | Problem                                                  | What to do                                                                            |
| ---------- | -------------------------------------------------------- | ------------------------------------------------------------------------------------- |
| Noise      | The value is different from the recommended marker.      | Check whether the custom value is intentional. Test the affected module.              |
| Obstacle   | A generative AI response times out too quickly.          | Increase the timeout for the related module if the expected response needs more time. |
| Obstacle   | A generative AI response takes too long.                 | Review whether the timeout is too high for the expected user experience.              |
| Obstacle   | Knowledge triggers the Not Expected flow after waiting.  | Review the Knowledge timeout value and test the affected Knowledge behavior.          |
| Obstacle   | Rephrase Answer returns a static answer.                 | Review the Rephrase Answer timeout value.                                             |
| Obstacle   | New intents cannot be imported or created for Zero-Shot. | Check whether the Zero-Shot token limit has been reached.                             |


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs-conversational-ai.syntphony.com/configurations/value-ranges.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
