Chats
How to work with models in chat: model selection, file upload, web search, artifacts, prompt library, voice input, sharing and Quick Actions.
The Chat is the central tool of the meinGPT platform. You type a question or task, the selected model answers, and with a few clicks in the input area you extend the chat with file uploads, web research, artifacts or saved prompts from the library.
Chats are private. Only you see your own chats. The meinGPT team has no access either, no automatic sharing happens.
Core capabilities in the input area
Model selection
Per chat, you choose the AI model that should answer the request. Depending on your workspace configuration, models from OpenAI, Anthropic, Google, Mistral and others are available. An EU flag next to a model indicates that it is hosted in an EU region and is GDPR-compliant. Reasoning models (with "Thinking" or "Reasoning" in the name) deliver more thorough results for complex tasks but take longer and consume more credits.
More on the model selection and which model fits when: AI Models.
File upload and sources
Use the paperclip icon to add files or sources directly to the chat. The content is sent along to the model so it can read, summarize or analyze the file.
For plain text files (PDF, Word, txt) the text is extracted, layout and formatting are lost in the process. For spreadsheet analysis with precise calculation results, the Excel agent is more reliable because it opens the file in the Code Sandbox instead of reading it as text only.
Insert a prompt from the library
Use the library icon to insert a saved prompt from the Prompt Library straight into the input area. If the prompt has variables, the library asks you for the values and substitutes the placeholders automatically. Useful for recurring requests where the structure stays the same.
Web search
With the web search toggle, you allow the model to research on the internet. It formulates its own search queries, loads the relevant pages and answers based on the found sources. Sources appear with a preview (title, description, favicon) so you can immediately tell what the answer is based on.
More details (number of sources, passing a URL directly, prompting tips): Web Search.
Artifact mode
Longer texts, reports, email drafts or other structured content shouldn't disappear into the chat history but should be treated as their own document? That's what the artifact mode is for, three settings (Auto, Always, Off). Active artifacts appear as a separate panel next to the chat and can be downloaded as PDF or Word.
More details: Artifacts.
Voice input (microphone)
Click the microphone icon to dictate your prompt instead of typing it. The language is detected automatically, transcribed, and inserted into the input field. Handy for longer prompts, when multitasking, or when typing isn't practical.
Share a chat
Activate the share toggle and generate a link to send to colleagues or to support. Recipients see the chat exactly as you shared it. Useful for best-practice libraries within a team or for support tickets where the agent needs to see the full conversation.
Comfort features
Input draft persistence
A prompt you haven't sent yet stays in the input field even after a page reload. The draft is stored locally in the browser. Particularly useful for long prompts or when F5 was hit by accident or the tab was closed.
Sources with preview
When the model uses web search, the sources appear with title, description and favicon. You can tell at a glance whether a source is trustworthy, without having to open it.
Rotating placeholder suggestions
The empty chat input shows rotating placeholder suggestions like "Summarize this PDF" or "Build me an Excel breakdown". They lower the entry barrier for new users and give inspiration when you don't know where to start.
Quick Actions
Quick Actions are customizable buttons that appear under a chat response and apply a predefined follow-up prompt to it with a single click. Instead of typing "Make this shorter" every time, you click More Concise and the assistant shortens the response.
Default Quick Actions
New organizations automatically receive the following Quick Actions:
- More Concise: makes text more precise and to the point
- More Detailed: expands text with additional detail
- Improve: general text improvement
- More Formal: converts text to a more formal tone
- More Casual: converts text to a casual tone
Managing custom Quick Actions (admin)
Workspace admins can create their own Quick Actions for tasks that come up repeatedly in a team. Examples: "Translate to English", "Turn this into an action list", "Draft a reply for an email request".
Navigate to Settings → Quick Actions
Click Add Quick Action
Fill in the fields:
- Label: the text shown on the button
- Prompt: the actual prompt text sent to the model
- Icon: pick from the available icons
- Enabled: toggle to activate or deactivate the action
Click Save
Edit, reorder, delete your Quick Actions
- Edit with the pencil icon next to an action: change label, prompt, icon or status, then Save.
- Reorder with drag & drop in the settings dialog. The new order shows up in the chat interface immediately.
- Delete with the trash icon and confirmation in the modal dialog.
Limits and permissions
- Maximum enabled actions: 8 per organization. You can create more, but only 8 can be active at the same time.
- All workspace members can see and use Quick Actions, only workspace admins can manage them.
Best practices for Quick Actions
- Be specific: a Quick Action with a clear prompt produces more consistent results
- Keep it short: concise prompts are easier to maintain
- Test before activating: run a Quick Action yourself once before the whole team sees it
- Use descriptive labels: the button text must be understandable in one second