Performance

What makes Aimogen Pro slow, which parts are inherently slow, and the settings that matter for a responsive site.

Aimogen Pro does two very different kinds of work, and they have different performance characteristics.

Background work — bulk generation, editing, agents, workflows — is slow by nature, because provider calls take seconds and there are many of them. Speed here means throughput and not timing out.

Front-end work — the chatbot, forms, generators — is felt directly by visitors. Speed here means latency.

Cron and background processingMaking scheduled work complete reliably. Large workloadsScaling bulk generation and editing.

What the plugin does not slow down#

On a page with no Aimogen shortcode, the plugin adds normal WordPress plugin overhead: its hooks load, and that is all. It does not make AI calls on page load.

The exceptions, all avoidable:

An [aicontent] shortcode with cache_seconds="0", which regenerates every view
A chatbot rendering on a page — the widget loads, but no AI call happens until a message is sent
Snippets, which run like any other code

Front-end latency#

For a chatbot, three things dominate:

1. The model. The largest factor by far. A flagship model can take several times as long as a fast mid-tier one. For chat, choose for latency. See Selecting models.

2. Prompt size. Every retrieved chunk, every message of history and every internet result is more input to process. Reduce the retrieval count and the context window.

3. Internet access. A search plus several page fetches before the model even starts. Turn it off for the chatbot unless you need it.

Streaming does not make responses faster, but it makes them feel much faster because text appears as it is generated. Keep it on unless you need retries, which do not work in streaming mode.

Caching#

Exclude admin-ajax.php from page cache. Otherwise visitors receive each other cached chatbot responses.

Object caching helps. The Snippets Engine caches its active list for 60 seconds in the aimogen group.

[aicontent] caches for 30 days by default. Leave it.

OmniBlock result caching avoids re-calling the AI for identical inputs.

More detail in Recommended server configuration.

Admin performance#

Some plugin screens are large. The Settings and Chatbot screens post many hundreds of fields, which is why max_input_vars matters: too low and settings silently fail to save.

The Live AI Activity screen polls while open. Close it when finished.

Database growth#

The logs tables grow with usage, and aiomatic_logmeta fastest when detailed tracking is on. A large wp_options table also slows every request, and the plugin stores several large options.

Set log retention, and see Maintenance and cleanup.

Still stuck? Open a support ticket and include the diagnostics from Aimogen Pro › System & Logs › System Info.