Understand the Verics interface
A calm tour of the main areas and controls in Verics, what each one does, and when you would reach for it. Some controls depend on your plan and configuration.
Navigation
Interface overview
Verics is built around a conversation. You type a task, the AI responds, and a set of controls around the message field lets you shape how that response is produced. The main areas you will notice are:
- a conversation area where your messages and the AI responses appear
- a message input where you type or dictate your task
- a model selector for choosing which AI model answers
- an agent selector for choosing a preset assistant
- a knowledge base option for using your own saved material
- a web search option for looking things up online
- file and image upload for attaching source material
- voice input for dictating a message
- a prompt library for saving and reusing prompts
- a compare option for answering with several models at once
- folders for organizing your conversations
- settings, where you manage your configuration
Not everything is always visible
Key controls explained
Each control below follows the same short pattern: what it does, when to use it, and one thing to keep in mind.
Model selector
What it does: chooses which AI model answers the current task. Models are grouped and labeled with a plain-language category such as Fast, Balanced or Deep analysis. When to use: whenever a task is noticeably simpler or more demanding than usual. Keep in mind: which models are available, and what they can do, depends on your configuration.
Agent selector
What it does: lets you pick a preset assistant configured for a particular kind of work. When to use: when a ready-made assistant matches your task better than a general conversation. Keep in mind: the available agents depend on your setup.
Knowledge base
What it does: lets the AI use your own saved material as context for its answer. When to use: when the task relies on documents or information you have added. Keep in mind: it only helps if you have added relevant material first.
Web search
What it does: lets the AI look up information from the web for its answer. When to use: for topics that may need current or external information. Keep in mind: still check the sources for anything important. A more thorough research option may be available depending on your plan.
File and image upload
What it does: attaches source material to your message. Images, PDF, Word and text files are supported. When to use: when you want the AI to work from a specific document or picture. Keep in mind: there is a file-size limit, and some binary formats cannot be read.
Voice input
What it does: lets you dictate a message with your microphone; it is transcribed into text. When to use: when speaking is faster than typing. Keep in mind: your browser will ask for microphone permission.
Prompt library
What it does: saves prompts you use often so you can reuse them. When to use: when you repeat similar tasks and want a consistent starting point. Keep in mind: a saved prompt is a starting point you can still edit before sending.
Compare mode
What it does: sends one prompt to several models and shows the answers side by side. When to use: when you want to see how different models handle the same task. Keep in mind: answering with several models uses more of your available usage.
Folders and folder memory
What it does: organizes conversations into folders and can optionally let a chat use other chats in the same folder as extra context. When to use: to keep related work together. Keep in mind: folder memory is optional and off unless you turn it on.
Reasoning effort and context limit
What it does: for models that support it, reasoning effort adjusts how much step-by-step work the model does; the context limit controls how many earlier messages of the current chat are sent along. When to use: to trade speed against depth, or to keep long chats focused. Keep in mind: not every model uses a reasoning setting.
Image and video generation
What it does: creates images from a text description, and video generation may be available depending on your plan. When to use: when you need a visual rather than text. Keep in mind: availability of these modes depends on your configuration.
Settings and API keys
What it does: is where you manage your configuration; depending on your setup you may add your own provider API keys and adjust parameters. When to use: during setup, or when adjusting how the product behaves. Keep in mind: the exact options depend on your account.
A recommended first workflow
Bringing the controls together, a dependable routine looks like this:
- Start a new conversation and describe your task clearly.
- Provide context — type it, or attach a supported file.
- Select a suitable model, or an agent if one fits the task.
- Turn on the knowledge base or web search only if the task needs them.
- Send, read the answer, and refine it with a short follow-up.
Common beginner mistakes
- Giving a vague task and expecting a precise answer.
- Leaving out context the AI would need to get it right.
- Using a heavy model for a trivial task, or a light model for a complex one.
- Turning on web search or the knowledge base when the task does not need them.
- Accepting the first answer without checking it.
Small steps beat big ones
Related learning
See it in the product
The fastest way to learn the interface is to open Verics and try one small task.