shimmy 1.8.1

Lightweight sub-5MB Ollama alternative with native SafeTensors support. No Python dependencies, 2x faster loading. Now with GitHub Spec-Kit integration for systematic development.
Documentation

shimmy

There is very little structured metadata to build this page from currently. You should check the main library docs, readme, or Cargo.toml in case the author documented the features in them.

This version has 11 feature flags, 2 of them enabled by default.

default

huggingface (default)

This feature flag does not enable additional features.

llama (default)

apple

coverage

fast

full

gpu

llama-cuda

llama-opencl

llama-vulkan

mlx

This feature flag does not enable additional features.