> For the complete documentation index, see [llms.txt](https://docs.aisera.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.aisera.com/aisera-platform/tenant-setup/aisera-platform-configuration/tenant-configuration-settings/hyperflows.md).

# Hyperflows

The **Settings > Configuration > Hyperflows** configurations control execution behavior, prompt handling, and response delivery for Hyperflows in your Aisera tenant. These are tenant-level settings that apply to all bots you create in this Aisera tenant.

### Enable Hyperflow V1

| **Type**    | Checkbox |
| ----------- | -------- |
| **Default** | Enabled  |

Controls which processing path Hyperflow uses for non-streaming LLM requests. Disable this only if Aisera Support instructs you to.

{% hint style="info" %}
This flag is scheduled for removal in a future release.
{% endhint %}

### Hyperflow Execution Model

| Type     | Dropdown                     |
| -------- | ---------------------------- |
| Default  | `GPT-4o`                     |
| Requires | Enable Hyperflow V1 disabled |

Sets the default large language model (LLM) for Hyperflow execution and testing across your Aisera tenant. Individual applications can override the tenant default when needed. Available models:

* **GPT-5.2** — Recommended model for Hyperflow execution. Not available in European deployments.
* **GPT-5.1** — Strong reasoning capabilities.
* **GPT-4o** — Reliable performance for standard use cases.

{% hint style="info" %}
This feature is in beta.
{% endhint %}

### Stream Hyperflow Response Outputs

| **Type**    | Checkbox |
| ----------- | -------- |
| **Default** | Disabled |

Controls whether Hyperflow delivers responses progressively as they are generated, including intermediate status updates and output from any sub-workflows it invokes. When disabled, all output is held until execution completes and Aisera silently drops intermediate status messages. Enable this for complex Hyperflows where showing users progress while they wait is beneficial. Requires a streaming-capable channel.

See also: [Enable Streaming Flow Response](#enable-streaming-flow-response)

### Max Execution Steps In a Hyperflow

| **Type**    | Integer |
| ----------- | ------- |
| **Default** | 10      |

Configure the maximum number of steps a Hyperflow runs before the Aisera Platform stops execution and returns a message to the user. Each step represents one full cycle: the LLM selects a function (API, Action, or Workflow), the function executes, and the LLM receives the result to make the next decision. When the limit is reached without a final answer, the Aisera Platform stops execution. Increase this value for complex, multi-step reasoning chains that require more than `10` function calls. Lower this value to reduce latency and token consumption for simpler flows, or to prevent runaway executions caused by poorly designed tool loops.

{% hint style="info" %}
The default of `10` applies per execution, not per conversation turn.
{% endhint %}

{% hint style="warning" %}
No system-enforced upper bound exists. Higher values increase both latency and token consumption per request. When the limit is reached, Aisera does not retry and the user must re-initiate the conversation.
{% endhint %}

### Sanitize Hyperflow Prompt

Strips HTML from a Hyperflow's configured prompt before it is sent to the LLM on every execution. When enabled, Aisera converts the prompt from HTML to plain text, preserving basic formatting tags like `<b>`, `<em>`, and `<br>` as markdown equivalents and removing all other HTML. Aisera uses this cleaned prompt as the LLM system message and as the input to injection detection scanning. Leave this enabled in most cases: Hyperflow prompts are typically authored in a rich-text editor that produces HTML, and LLMs perform better with clean plain-text system prompts. Disable only if your prompt intentionally contains HTML that the LLM is expected to interpret, or if the conversion is degrading your flow's performance.

{% hint style="info" %}
Disabling this setting also affects the prompt passed to injection detection, which may reduce its accuracy.
{% endhint %}


---

# 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 dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.aisera.com/aisera-platform/tenant-setup/aisera-platform-configuration/tenant-configuration-settings/hyperflows.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
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.