Download Termite
Termite is a lightweight ML inference service for embeddings, reranking, and chunking. It's similar to Ollama but designed for non-generative AI models.
Termite ships with bundled ONNX Runtime and XLA backends for optimized local ML inference without requiring external model runtimes.
Installation Instructions
brew install --cask antflydb/antfly/termiteor tap homebrew-antfly and install separately:
brew tap antflydb/antfly
brew install --cask termiteThis installs Termite with bundled ONNX Runtime and XLA libraries for local ML inference.