local-llm

TypeScript multi-agent orchestration framework. Describe a goal, a coordinator decomposes it into a task DAG that runs on any LLM: Claude, ChatGPT, Gemini, DeepSeek, or local models.

Run Kimi K3, a 2.8T-parameter Mixture-of-Experts LLM, on a single Apple Silicon Mac. Streams MXFP4 experts on demand over HTTP into a local disk cache — fused NEON kernels, Metal/MPS compute, exact reproducible decoding, and an OpenAI-compatible API server for local chat and coding agents.

The free open source Grammarly alternative offline writing assistant assist with grammar, rewriting & tone entirely on your machine.

An AI-native OS. Models run on your own hardware, nodes federate peer-to-peer with no broker, and the API is OpenAI-compatible. Boots, has its own Wayland compositor, and runs on 8GB. Apache 2.0.

Writing that's kin to yours

HotPin: routing-guided expert pinning for lossless low-RAM MoE inference on consumer CPU+NVMe - 120B model at 3.84 tok/s in 19 GB RAM, no GPU