Ollama

Ollama 0.30 is now available with improved performance and GGUF model compatibility, bringing support to more models on a wider range of hardware. Performance on NVIDIA hardware is now up to 20% faster, and Vulkan is enabled by default for wider GPU acceleration.

OpenJarvis is an open-source framework for building personal AI agents that run on local hardware, developed by Stanford's Hazy Research and Scaling Intelligence labs. It's now available with built-in support for Ollama, allowing users to run local models and track energy, cost, and latency.