MindWorkAI/AI-Studio

MindWork AI Studio – What it is

MindWork AI Studio is a free, cross‑platform desktop application (macOS, Windows, Linux) that lets you talk to large language models (LLMs) from many providers – OpenAI, Anthropic, Google Gemini, Hugging Face, self‑hosted llama.cpp/ollama, and dozens of others. It acts like a smart chat client that can also run assistants (pre‑built workflows such as translation or summarisation) and tools (web search, Confluence lookup, etc.) without you writing prompts.

---

Core capabilities (as described in the README)

Category	What the app does
Model access	Connects to any LLM you have an API key for, including the latest OpenAI GPT‑6/5, Anthropic Claude, Google Gemini, Groq, DeepSeek, Qwen, Mistral, and open‑source models via llama.cpp, Ollama, LM Studio or vLLM.
Local RAG (beta)	Indexes files in your own folders (using Qdrant) and lets the model retrieve relevant passages, with a built‑in semantic‑search option.
Assistants	Ready‑made, no‑prompt tasks (e.g., translate text, generate slides, document analysis).
Tool use	Models can call external tools on their own – web search, read a web page, query Confluence, and future plugins.
Voice & transcription	Record audio/video, send it to a selectable transcription provider (including self‑hosted options) and get a text transcript.
Export	Chat answers can be saved as Word, OpenDocument, LaTeX, Markdown, or a full HTML page; tables are exported as spreadsheets.
Enterprise management	IT admins can centrally roll out providers, assistants, chat templates, and enforce data‑privacy policies across an organization.
Low‑resource friendly	Claims to run on cheap hardware (e.g., a €100 Raspberry Pi) with modest memory and storage footprints.
Open‑source build	Source can be compiled (see documentation/Build.md) and extended via a plugin system for tools, assistants, and model‑capability definitions.

---

Who might use it?

• Individual power users who want a single UI for all their LLM subscriptions and local models.
• Small teams or families looking for a free, cross‑platform chat client that respects privacy and lets you bring your own API keys.
• Enterprises that need centralized control over which models/tools employees can use and want to enforce data‑security policies.
• Developers interested in extending the platform with custom assistants or tool plugins.

---

Getting started (from the README)

1. Download the installer for your OS from the Setup page linked in the README.
2. Add a provider – enter an API key for the LLM you want to use (OpenAI, Anthropic, etc.) or point the app at a local server (llama.cpp, Ollama, vLLM, etc.).
3. Pick an assistant or start a normal chat. You can also enable tools like Web Search in the chat settings.
4. (Optional) Enable local RAG – point the app at a folder, let it build an index, and ask the model questions that draw on your own documents.

---

License

The project is released under a Functional Source License (FSL‑1.1‑MIT). You may use, modify, and share the software for personal, educational, or research purposes, but you may not create a commercial product that competes with MindWork AI Studio. Each release automatically converts to a permissive MIT license two years after its publication.

---

Community & support

• Funding – The maintainers accept monthly or one‑time sponsorships via GitHub Sponsors; contributors are listed in‑app.
• Roadmap – Planned items include deeper data integration (PDF/Office/Markdown), enterprise data retrieval (ERI), more assistants, a writing‑mode, voice‑output, browser extensions, and email‑monitoring.
• Documentation – Setup, build, tool development, and model‑capability guides are provided in the documentation/ folder.

---

Bottom line

MindWork AI Studio is a genuine, open‑source desktop client that aggregates many LLM providers, adds RAG and tool‑use capabilities, and offers both a consumer‑friendly UI and enterprise‑grade management features. It is firmly within the AI/LLM software space and not merely a curated list or tutorial collection.