> ## Documentation Index
> Fetch the complete documentation index at: https://docs.genow.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# Choose the Most Suitable Large Language Model (LLM)
> Learn how to pick an AI model and a response mode for your question.
This article explains where to find the model selection, which Gemini models are available, and how to choose a response mode.
***
## What an LLM Does
A Large Language Model (LLM) is an advanced AI trained on vast amounts of text to understand and generate human-like language. Genow uses an LLM as the final step: it takes the relevant documents found by our search system and turns them into a clear, conversational answer.
Our platform is **model-agnostic**, meaning it can work with various LLMs, but we provide Google's powerful Gemini models as the standard offering.
***
## Where to Select Your Model
The model selection sits directly in the input field, next to where you type your question. Click the model name to open the list of models your administrator has made available and select the one you want to use.
> [!NOTE]
>
> The model selection used to sit at the top of the user interface. It moved into the input field so that the model and the response mode can be set in the same place.
***
## Response Modes
Next to the model selection in the input field, you can also choose a response mode. The response mode tells Genow how much effort to put into your answer.
| Mode | When to use it |
| - | - |
| Fast | Everyday questions, quick lookups and short summaries. This is the default and the right choice for most tasks. |
| Balanced | Questions that need a little more reasoning, but where you still want a quick reply. |
| Precise | Complex questions, detailed analysis, or answers that have to be as accurate as possible. |
> [!NOTE]
>
> Start with Fast. Only switch to Balanced or Precise if an answer is not detailed or accurate enough for your task.
***
## Available Gemini Models
> [!NOTE]
>
> **Always pick the newest model available to you.** Each generation improves on the one before it, so a current model usually gives you better answers *and* a faster response than an older one - no matter whether the older one was a Flash or a Pro model.
The newest model on the platform is **Gemini 3.8 Flash**. It delivers higher-quality answers in a shorter amount of time and is the right choice for the large majority of questions.
You will also see older models in the list. The names follow a simple pattern:
* The **number** tells you the generation. A higher number means a newer model.
* **Flash** is optimized for `speed and efficiency`, **Pro** for `deep reasoning`. This distinction only matters between models of the same generation - a newer Flash model will normally outperform an older Pro model.
> [!NOTE]
>
> If you need more depth on a complex question, change the response mode to Balanced or Precise before you switch to an older model. The response mode is the better lever for this.
This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.