Changelog

GPT-6.1 Terra is now available in innoGPT and unlocked directly in the model selector for everyone. The new OpenAI model responds three times faster than its larger sister models and excels at analyzing images, screenshots, and documents.

You can complete daily tasks like reading invoices, reviewing technical drawings, or drafting proposals in seconds while saving half the token costs compared to its predecessor. Simply select GPT-6.1 Terra at the top of the chat.

Have fun trying it out! 🙌

Further Improvements

• You can now connect a local folder from your device in project chats within the desktop app, create or edit Word and Excel files directly inside it, and run Python scripts on it.

• Local commands in the desktop app now run via PowerShell on Windows.

• For local MCP servers in the desktop app, you can now decide for each individual tool whether it is allowed directly, asks for confirmation beforehand, or is blocked.

• MCP connections remain stable across multiple messages, new tools are ready to use immediately, and searching for connections no longer times out on the first load.

• When removing an MCP server, you can now see in advance exactly who will lose access.

• All new teams reliably start with the official skills, and all templates for skills, assistants, workflows, and knowledge folders are available in German and English.

• You can add a delegation description to assistants so the main chat routes requests directly to the right specialist assistant.

• Your team home page loads faster to reduce waiting times.

• Running agent tasks are marked with the status "Running" in the sidebar across all devices.

• Interrupted chats resume more reliably and messages are no longer accidentally saved twice when sent.

• Chats, scheduled tasks, and system notifications now consistently respond in the language you write in.

• Scanned and password-protected PDFs are processed without errors, and project files no longer get stuck during upload.

• The "Improve prompt" feature automatically includes attached files in its optimization.

• When multiple assistants are selected, the input field shows all matching actions again, the Outlook integration responds as usual, and unknown tool calls no longer interrupt the conversation.

• Unreadable output files no longer block further downloads, and scheduled tasks without results no longer send empty notifications.

• Shared pages remain directly visible to logged-out visitors, and third-party sign-in in the desktop app no longer interferes with logging into innoGPT.

• In meeting notes, editing no longer accidentally opens the source citation, the preview card in the image model selector stays open reliably, and loading apps no longer displays a distracting splash screen.

Starting now, the new projects with their own team channel are available in every workspace. This makes AI in innoGPT a true multiplayer tool for your team. Your team discusses together in the channel, works from the same project knowledge, and lets innoGPT write finished results directly into your shared work folders.

Why we built this

Many teams asked for a way to share chats with each other. But someone else's chat history only shows one person's unfinished thinking process. Real collaboration needs a shared workspace where the team and the AI work on the same result. If your marketing team is planning a major customer event, for example, everyone coordinates directly in the project channel, in clear threads. If someone brings in @innoGPT, the AI immediately knows all the stored briefings, answers questions in the thread, and saves finished invitation texts or spreadsheets directly in the connected work folder for everyone.

What you can do with it

• Collaborate in the project channel: Exchange ideas with your team in threads in the #General channel, and bring in the AI as a team member when needed by mentioning @innoGPT.

• Save results in work folders: Connect up to five work folders from your Files page, where innoGPT and your team can read, edit and save new results together.

• Bundle knowledge sources: Upload documents or link folders from SharePoint, OneDrive and Google Drive. innoGPT searches these first for every question in the project.

• Build a shared memory: Save important project details in the settings or with "remember ..." in the chat, so the whole team and the AI stay on the same page.

• Set instructions and model: Define central rules for the entire project and choose the response model or custom assistants for the channel.

• Keep access clearly separated: Invite colleagues or the whole team to the shared channel, while your own individual chats within the project remain your private drafting space.

How to get started

Click "Projects" in the left sidebar and create a new project with a name, icon and colour using the button at the top right. The gear icon opens the project settings, where you invite your team and connect your sources and work folders. At the top of the project view, you can switch at any time with one click between your "Overview" for private individual chats and the shared "Channel".

We look forward to your feedback! 🙌

Gemini 4 Argon is now available in innoGPT. Google's new model is especially good at complex analysis, large amounts of data of up to one million tokens, and demanding code. You enable it in the model settings and then select it directly in the chat. You can find all the details about the model on the Google blog.

Workflows in Public Beta

Workflows are now in public beta and available in every workspace. You no longer need to request access first, so you can get started right away and automate your own processes.
We look forward to your feedback! 🙌

More improvements:

• innoGPT now loads small assistant knowledge collections into the context in full, so answers build on them even more precisely.

• The settings pages for chat, image and video models show all models clearly on one page, grouped on provider cards with clear hosting zones.

• In the desktop app, innoGPT now also reads PDFs from your local folder and runs more tool calls per answer.

• Documents in the chat are found more reliably by their file name and read in full.

• innoGPT renames attachments with identical names so each name is unique, and rejects ambiguous file names.

• The chat loads noticeably faster because the text editor is only loaded in the background when needed.

• Your meeting summaries give consistent, clearly structured results, no matter which language model you choose.

• Only a single recording bot now reliably joins a meeting.

• Searching within meeting transcripts no longer times out, even for long recordings.

• The page for editing meetings now has all German translations in place.

• In the model search, the Enter key no longer makes a selection while you are composing characters with an input method editor (IME).

• The share dialog for assistants now loads without errors, even when former team members have already left the workspace.

• Workflows that you pause through the API now correctly show as “Paused” in the interface.

• After you grant access in the browser, the Google Workspace connection reliably takes you back to innoGPT again.

• The connection to MCP servers stays stable when your sign-in is renewed in the background.