Working with Chat Responses

With ELITEA, you can review the LLM responses in chat, interact with them, and play back the whole conversation in chat.

​

How to Interact with Chat Responses

ELITEA allows interacting with responses generated by your AI assistant:

​

Send a Message While the Agent Is Working

Mid-turn input allows sending a message while an AI assistant is still generating its answer, to change the rest of the run instead of waiting for it to finish. If mid-turn input is allowed and turned on for your project, the message input field remains active while the LLM is still generating its response. This means you can enter your text in the message field as usual. After you select Send, ELITEA will inject your message into the current context and adjust its response based on your new input instead of starting a new run. ELITEA automatically sends your text as a new, separate message instead if:

• It exceeds 4000 characters.
• The running turn ends before ELITEA processes it.

The text of your mid-turn input is labeled as sent while running and appears indented under your original message.

​

Edit Your Latest Message

If you want to refine your prompt instead of sending a new follow-up message, you can edit your latest user message and regenerate the answer from the updated text:

1. Among the action icons shown under your latest user message, select Edit (the pencil icon).
2. The message switches to inline edit mode. If the original message included attachments, they are not removed and remain visible.
3. Update the text, as necessary.
4. (optional) Add new attachments or remove the existing ones.
5. Select Save and apply to submit the edited message and regenerate the answer, or select Cancel to discard your changes.

The Edit icon appears only under your latest message, and only after the response was generated. The Save and apply button stays disabled until you add any changes to the message. Editing a message permanently replaces the previous response — ELITEA does not keep a history of earlier responses, and this action cannot be undone. Regenerating counts toward your project’s model and token usage like any new message.

​

Read Out a Response

Read out uses text-to-speech to play an AI response aloud instead of you reading it:

1. Under an AI response, select Read out.
2. ELITEA opens a floating voice player under the message input.
3. Select Play in the voice player to start listening.

While a response plays, ELITEA highlights the part of the text currently being spoken. ELITEA adapts some content before speaking it instead of reading it literally:

• Code blocks, tables, and diagrams are replaced with a short spoken notice.
• Links are read as “the link” (or by their link text, if descriptive).
• Images are read as “an image showing” their alt text.
• Ordered list items are read with spoken ordinals (for example, “First,” instead of “1.”).

​

Copy a Message to the Clipboard

You can copy the text of any message — yours or an AI response — to your clipboard:

1. Under the message, select Copy to clipboard.
2. ELITEA copies the message and shows a confirmation.

ELITEA copies the message’s raw content, not its rendered formatting. For a message that includes canvas content, attachments, or an error, ELITEA copies the canvas text, an attachment name placeholder, or the error details instead of the surrounding message text.

​

Regenerate the Latest Output

You can request a new answer for the same input without editing your message, once ELITEA has generated at least one response in the chat:

1. Under the latest generated response, select Regenerate.
2. ELITEA regenerates the output based on the same input, providing a refined or corrected response.

You cannot regenerate the latest output while a response is still streaming, while you are editing a canvas in the chat, or if the AI participant that produced the response is no longer active in the chat. Regenerating resets the response and its reasoning trace — ELITEA does not keep the earlier version, and this action cannot be undone. Regenerating also counts toward your project’s model and token usage like any new message.

​

Delete Your Latest Message

You can delete the latest message in the chat — either your latest message or the AI’s latest response.

1. Under the latest message, select Delete.
2. Select Delete again in the confirmation dialog, or Cancel to keep the message.

Deleting a message removes only that message — it does not remove any other messages, related or not. ELITEA does not keep a copy of a deleted message, and this action cannot be undone. You cannot delete the latest AI response while you are editing a canvas in the chat or while the response is still processing.

​

How to Use Playback Mode

In playback mode, you can step through the conversation in the current chat turn by turn, exactly as it happened, without sending any requests to the models or toolkits:

1. Find the necessary chat in the CHATS sidebar.
2. In its overflow menu, select Playback.
3. Move forward to the next message, go back to the previous message, or stop the playback simulation using the on-screen controls or keyboard arrow keys.

Benefits of Playback Mode for Demos During demos, teams can use playback mode to demonstrate their collaboration and decision-making processes to stakeholders:

• You can demonstrate workflows and complex interactions without incurring processing costs or waiting for live responses.
• Stakeholders see the exact flow of discussions and decisions smoothly and without interruptions.
• The step-by-step simulation keeps your audience engaged and focused on key points.