Privex AI
A unified, private space for cloud API keys and local models.
To recommend the best model for your device, we need to check your hardware.
CPU core count, approximate RAM, and GPU name from your browser.
Send this data anywhere, store it on a server, or share it with anyone.
Entirely on your device. Analyzed locally. Deleted after you close this screen.
Hardware Detection
Reading local capabilities
Confirm Specs and Use Case
Ensure these values are accurate for the best recommendation.
Limited Hardware Detected
Your device has limited memory. We recommend using Cloud API mode instead of downloading local models, as performance will be extremely poor.
Llama 3.1 8B
Excellent quality, versatile. The gold standard mid-size model.
- Install Ollama for your OS:
Windows macOS LinuxRun the downloaded installer.Unzip and drag to Applications.
curl -fsSL https://ollama.com/install.sh | sh - Pull the model:
ollama run llama3.1Downloads ~. Make sure you have a good connection. - Come back here and click Connect.
You already have models installed!
We ranked your installed models based on your hardware. Select one to use.
Ollama not found on your system
We couldn't reach your local model server. Follow this guide to get running.
Download Ollama
Download and install Ollama for your operating system from ollama.com/download.
System Spec Analyzer
Enter your hardware specs to get a model recommendation:
Connect Ollama
Run Ollama on your machine. If you are using the web version, you must allow browser connections first:
(Restart Ollama after running the command)