PromptQuorum Desktop App: Public Beta Is Now Live
PromptQuorum's desktop app is out of private testing and available to download for macOS, Windows, and Linux.
Quick Answer
Quick Answer
- ✓PromptQuorum's desktop app left private beta on July 13, 2026 and is downloadable for macOS (Apple Silicon and Intel), Windows 10/11, and Linux (x64 and ARM64).
- ✓The build is unsigned during this beta phase, so macOS and Windows will show a security warning on first launch — that is expected, not a sign of a compromised file.
- ✓The app needs an AI backend to run prompts: install Ollama or LM Studio for local models, or bring your own API key (BYOK) for OpenAI, Anthropic, or another provider.
- ✓Model agreement is not proof of correctness — verify factual claims independently before relying on multi-model consensus output.
- ✓Get the current build from the download page and send bug reports directly to the team by email.
What Is the PromptQuorum Desktop App?
The PromptQuorum desktop app dispatches one prompt to multiple AI models at once and scores how closely those models agree — a process PromptQuorum calls "Quorum." It runs as a native application on macOS, Windows, or Linux, separate from the web-based prompt engineering guides on this site.
The app supports two ways to run models: a local AI engine (Ollama or LM Studio) on your own machine, or cloud providers through your own API key. No PromptQuorum account is required during this beta.
Which File Do You Need?
Match your operating system and chip to the right installer before downloading.
| Platform | File | Notes |
|---|---|---|
| macOS — Apple Silicon (M1–M5) | .dmg installer | Mount the disk image, drag the app into Applications |
| macOS — Intel | .dmg installer | Mount the disk image, drag the app into Applications |
| Windows 10 / 11 | .zip archive | Extract the archive first, then run the .exe inside it |
| Linux — x64 (Intel/AMD) | .AppImage | chmod +x, then run directly — no install step |
| Linux — ARM64 | .AppImage | For Raspberry Pi 5 and other ARM64 devices |
Not sure which Mac you have? Open the Apple menu, choose About This Mac, and check the "Chip" or "Processor" field.
Linux support is newer than macOS and Windows support and has seen less real-world testing. If the app is blocked on Ubuntu, try launching it from a terminal with the --no-sandbox flag and report what you see.
Verify Your Download (Optional but Recommended)
Each release ships with a SHA-256 checksum on the download page so you can confirm a file arrived complete and unmodified before you install it.
- •macOS: open Terminal and run shasum -a 256 ~/Downloads/[filename]
- •Windows: open PowerShell and run Get-FileHash [filename] -Algorithm SHA256
- •Linux: open a terminal and run sha256sum ~/Downloads/[filename]
Set Up a Local AI Engine Before First Launch
PromptQuorum works best with a local LLM engine running on your machine — this keeps prompts private and avoids per-token API costs. Skip this section entirely if you plan to use a cloud provider with your own API key instead.
- •Ollama (simplest first setup): download it from ollama.com, install it, then pull a model from a terminal, for example ollama pull gemma3:1b. It runs in the background and PromptQuorum detects it automatically.
- •LM Studio (more GUI-driven): download it from lmstudio.ai, browse and download a model inside the app, then start the local server from the Developer tab.
- •Cloud only: skip local setup and add an API key for OpenAI, Anthropic, or another provider (bring your own key) during onboarding.
What Happens on First Launch
macOS: right-click (not double-click) the app in Applications and choose Open, then confirm Open in the dialog warning the app cannot be verified. If macOS still blocks it, go to System Settings, then Privacy & Security, and choose Open Anyway.
Windows: Microsoft Defender SmartScreen shows "Windows protected your PC." Click More info, then Run anyway. Windows remembers this choice for later launches of the same file.
Both warnings are expected for this beta — the app does not yet have a paid code-signing certificate and will stop showing them once signed builds ship.
If you installed Ollama or LM Studio, keep it running before opening PromptQuorum — the app checks for it on startup and prompts you to configure an AI provider if none is detected. No account creation is required for this beta.
Model Agreement Is Not Proof of Correctness
PromptQuorum scores how closely multiple AI models agree on a response. High agreement means the models converged on the same answer — it does not mean the answer is factually correct, since every model in a dispatch can share the same training-data blind spot or bias.
Independently verify factual claims, numbers, and citations before relying on them. Do not use outputs for medical, legal, financial, or safety-critical decisions without independent verification. Hallucination detection and consensus scoring are still being refined during this beta.
Report Bugs Directly to the Team
Send feedback to hanskuepper@promptquorum.com with what you were trying to do, what happened versus what you expected, your operating system and app version, and a screenshot if possible. Bugs, confusing UI, crashes, and features that do not work as described are all useful reports.
The Desktop App Is Half the Picture
The app is the workspace; the [PromptQuorum blog and guides](/blog) are the reference. They cover local LLM setup, Ollama and LM Studio comparisons, and model selection in nine languages — useful reading before or after your first install.
Download the Beta
Pick your platform, verify the checksum, and install — private beta, unsigned build, free to use.
↓ Go to the Download PageFrequently Asked Questions
Is the PromptQuorum desktop app free?+
Yes. It is free to use during the beta. Local AI models run at no per-token cost; cloud providers billed through your own API key follow that provider's pricing.
Which operating systems does the desktop app support?+
macOS (Apple Silicon M1–M5 and Intel), Windows 10 and 11, and Linux x64 and ARM64 via AppImage. Linux support is newer and has seen less real-world testing than macOS and Windows.
Do I need Ollama or LM Studio to use PromptQuorum?+
Yes. PromptQuorum needs an AI backend to run prompts — Ollama or LM Studio is the easiest way to run one locally, but you can also use your own API key for a cloud provider (OpenAI, Anthropic, etc.) instead.
Why does my operating system warn that the app cannot be verified?+
The beta build is not yet code-signed, so macOS shows an unverified-developer warning and Windows SmartScreen blocks it by default. Both are expected and can be bypassed with Open Anyway (macOS) or Run anyway (Windows).
Is my data private when I use the desktop app?+
If you use a local AI engine, prompts stay on your machine and are not sent to PromptQuorum servers by default. If you use a cloud provider via API key, your prompts go directly to that provider under its own data policy.
How do I report a bug during the beta?+
Email hanskuepper@promptquorum.com with what you were trying to do, what happened, your OS and app version, and a screenshot if possible.