Announcing Prem — Private Open Source LLMs for ALL
Introducing Prem
Today we are excited to announce Prem, the world’s first Privacy-centric and Composable AI Platform — unveiling AI THAT YOU OWN!
Leveraging state-of-the-art Open-Source Large Language Models (LLMs), Prem provides a unified environment to develop AI applications and deploy AI models on your infrastructure. Abstracting away all technical complexities for AI deployment and ushering in a new era of privacy-centric AI applications — users can finally retain control and ownership of their models.
Context & Opportunity
The current AI landscape is largely dominated by centralized “big-tech” organizations controlling the world’s most advanced models. Most of these models are closed-source, cloud-hosted, expensive, and without verifiable privacy options.
In parallel, there has been a renaissance of innovation from the Open Source Community. Like the text-to-image Stable diffusion in 2022, which took the world by storm or the latest release of the Llama model from Meta which sparked a new wave of open-source development in the LLMs Arena. The rate of improvement has been unprecedented and open-source LLMs are quickly catching up in performance with their centralized rivals.
This encourages individuals and organizations globally to leverage AI’s capabilities without depending on large tech firms. However, incorporating open-source LLMs into applications remains a considerably more complex task than hosted solutions due to technical hurdles that keep the entry barrier high.
Privacy & AI
AI models, such as LLMs, learn human language from extensive datasets, which can include sensitive information. Developers heavily rely on cloud-based services to conduct inference and deploy these models, requiring organizations to transfer all their data to centralized third parties. These often lead to data exposure, data breaches, and unauthorized access. Moreover, there seems to be an incentive from these cloud providers developing LLMs to use the information shared in the applications as input to retrain these models. This raises privacy concerns about how these models are trained and the data sharing policy. That is why companies like Samsung, Apple, Verizon, JPMorgan, and others are restricting their employees from using these services.
Vision
Our vision is to engineer a world where individuals, developers, and businesses can embrace the power of AI without compromising their privacy. We believe in a future where users retain ownership of their data, AND the AI models trained on it.
Mission
Our mission is to accelerate the advent of privacy-centric and open-source AI models. To achieve this, we developed a powerful AI platform designed to deploy self-hosted AI models easily.
The Prem App
As a stepping stone toward our mission, we’re honored to introduce you to Prem App for MacOS. This fully open desktop application serves as your primary interface for using the Prem Operating System: from launching your self-hosted ChatGPT to your StableDiffusion model — all at your fingertips, with easy-to-integrate API out of the box.
Prem App can be downloaded as a Mac desktop application designed to run on your own computer without exposing sensitive data to third-party. Prem can be installed on any Linux server for technical users and production deployments requiring GPU.
On top of that, we envision an explosion of developers’ integrations and composability across a full array of the latest open-source models.
The Core Features of Prem:
🚀 One-Click Deployment: Launch your AI models with a single click.
🌐 Offline Operations: Seamlessly perform operations and access features without an internet connection.
🔒 Privacy-centric Design: Prioritize user privacy and data protection through a thoughtfully designed system.
🔌 Effortless Integration: Seamlessly implement machine learning models with the user-friendly API interface for LLM models.
🌍Ready for the Real World: Bypass the complexities of inference optimizations. Prem’s got you covered.
🏃♂️ Rapid Iterations, Instant Results: Develop, test, and deploy your application in minutes.
Install Prem Anywhere
We encourage users and developers to run their models without the hassle.
Running Prem on your Mac is great for non-technical users to get a glimpse of the capabilities of a full-fledged App Store of open-source AI models.
Running Prem on your servers: AI models require powerful GPUs and always-on servers. Prem App comes with a convenient Linux installer script to install the full PremOS on your own remote server.
wget -q https://get.prem.ninja/install.sh -O install.sh; sudo bash ./install.shMinimum requirements for serverse 📝
🔵 16GB RAM
💾 80GB free space
💻 Debian or Ubuntu OS
How to Get Started?
📥 Go to https://premai.io to download Prem App.
👀Can’t install Prem? Try the live demo here: https://app.prem.ninja/
Join Us
AI is inevitable. How we intereact with it is our choice. Join us to perserve our right to privacy in the age of AI.
Discord Community — https://discord.com/invite/WkBkzVEp
Follow us on Twitter — https://twitter.com/premai_io
Follow our Updates on GitHub — https://github.com/premAI-io/prem-app
