Change8

v0.30.0-rc17

📦 ollamaView on GitHub →
3 features

Summary

This pre-release updates the architecture to directly support llama.cpp, enabling GGUF compatibility and utilizing MLX for Apple Silicon acceleration. Feedback is requested on performance and memory utilization.

Migration Steps

  1. When installing on Mac/Linux, use OLLAMA_VERSION=0.30.0-rc15 in the install script: curl -fsSL https://ollama.com/install.sh | OLLAMA_VERSION=0.30.0-rc15 sh
  2. When installing on Windows, set the environment variable OLLAMA_VERSION to "0.30.0-rc15" before running the install script: $env:OLLAMA_VERSION="0.30.0-rc15"; irm https://ollama.com/install.ps1 | iex

✨ New Features

  • Architecture changed to directly support llama.cpp instead of building on top of GGML.
  • Added compatibility with GGUF file format.
  • MLX is now used to accelerate model inference on Apple Silicon.