When everything happens inside one large conversation, that conversation gradually accumulates unrelated topics: client questions, market research, workflow setup, document analysis, and product support.

Over time, the conversation history becomes diluted. When the AI generates a response, it must interpret a large amount of mixed context.

Threads solve this problem.

Each thread maintains its own conversation history and focus. This allows Vega to reason within a clearer context, which increases the likelihood of generating more precise and relevant outputs.

In practice, this means you can maintain separate threads for:

Each thread becomes a focused workspace for that topic.

This Help Center article focuses on threads in the Vega chat, but the same concept also applies to the Vega Outlook plugin and the Vega Gmail assistant.

In those environments, each email conversation you open in your inbox already functions as a thread in Vega. When you interact with Vega in that email conversation, Vega uses the email thread as the context for the discussion.

The Vega chat in the web platform allows you to create and manage multiple threads.

Each thread represents a focused conversation with Vega. You can return to a thread at any time and continue the discussion with the existing context already available.

This is particularly useful when working through multi-step tasks such as:

Instead of restarting the context each time, the thread keeps the conversation organized.

You can create a new thread by clicking New at the top right of the Vega chat.

Vega also supports Push to Talk (PTT), which allows you to dictate your message instead of typing it.

Simply press and hold the space bar on your keyboard to start dictation.

This feature is designed to make conversations with Vega faster and more natural, especially when you want to ask longer questions or explain context in detail.

You can open the Vega chat from anywhere in the platform.

You can open it in two ways:

1. Click the AI icon in the top right of your screen

This will open the latest thread you were using from the home page of the Vega web platform.

2. Click the (+) icon and select “New chat.”

This will open a new thread. If you are on a page such as a meeting page and clients are linked to that meeting, Vega will automatically link those clients to the thread.

When you open a thread from somewhere outside the dedicated Vega chat (for example from another page in Vega), Vega is aware of the page you are currently on. This context helps Vega interpret your request.

Threads are not only about organizing conversations. They also allow you to interact with several Vega capabilities directly from the chat.

You can link a client to a thread.

When a client is linked, Vega understands that the conversation relates to that specific person. This helps Vega retrieve the correct information when answering questions or creating records.

When you interact with Vega in a thread, different AI agents may be activated depending on the request.

These agents allow Vega to retrieve and reason across different types of information, including:

For a deeper explanation of how AI agents work in Vega, see this Help Center article.

You can attach files directly to a thread by clicking the paperclip icon at the bottom right of the Vega chat, next to the send button.

Vega can process and analyze those files as part of the conversation. This includes tasks such as:

Common examples include tax returns, portfolio reports, and other financial documents.

At the bottom of the Vega chat, just below the message area, you can choose between Instant and Thinking.

Instant is the default mode. It is designed for speed and works well for most day-to-day questions, retrieval requests, and quick back-and-forth conversations.

You can also switch to Thinking when you want Vega to spend more time reasoning before responding. This mode is better suited for requests that require deeper analysis, more synthesis, or more careful interpretation.

You can also ask Vega to create objects directly from the chat.

Depending on your setup, Vega can generate and prepare records such as:

These objects can then be reviewed and synced to your CRM, or stored directly in Vega if you are not using a CRM with Vega.

For example, you could tell Vega:

Here is a practical example of how a thread can be used as a complete workflow: