We're thrilled to announce the v1.4 release of Private LLM for macOS, designed to elevate your experience with a slew of powerful enhancements. This update focuses on improving performance, text generation, and user experience. Here's what you can expect:
- OmniQuant: A Quantum Leap in Text Generation: All models are now optimized using the cutting-edge OmniQuant quantization algorithm. This technological advancement significantly enhances the models' perplexity and text generation capabilities. In layman's terms, expect smarter and more coherent responses from Private LLM.
- Fresh Updates for WizardLM V1.2 Fans: If you've previously downloaded the 13B WizardLM V1.2 model, you're in for a treat. You can now opt for an improved version that leverages the new OmniQuant algorithm for even better performance.
- Special Option for Apple Silicon Macs: For those with Apple Silicon Macs boasting 16GB or more RAM, we now offer the option to download the advanced 13B parameter Wizard LM-13B-V1.2 model. Unleash the full power of Private LLM for a premium experience.
- General Improvements: Alongside these major features, we've also rolled out minor bug fixes and refinements to make your experience smoother than ever.