Decoding GitHub Repositories (2026) · Open Source Tools

A local inference engine for Apple silicon, built around the model.

Your iPhone helps your Mac run a 27B model: faster prompt reading and more context over a USB-C cable

llama.cpp fork for significantly improved performance on Ampere (especially RTX 3090 / 3090 Ti): example: 95+ tok/s over a 100K-token generation at temperature 1 for qwen3.8, supporting full 262K ctx.

Peer-to-peer LLM inference in the browser: pool your devices to run big open models, for chat and coding agents.

The set-and-forget LLM engine for Pascal and Volta: PXQ codec + kernels, auto-tuned per card. Ready-to-run PXQ models in MODELS.md; benchmarks vs llama.cpp in the README.

llama.cpp/ik_llama.cpp launcher: loads big MoE models across mismatched multi-GPU rigs by exact VRAM math.

ECMWF's GRIB and BUFR decoding/encoding library

I/O interface and utilities for CCSDS binary spacecraft data in Python. Library used in flight missions at NASA, NOAA, and SWRI

An open-source virtual AI Infra team that discovers, benchmarks, and safely upgrades local LLM inference—with independent quality gates, a stable OpenAI-compatible API, and automatic rollback.

JEV-inspired parallel decisions for CUDA LLMs. One context, many decisions. vLLM API, game-agent examples, and reproducible benchmarks.

Claude Code Monitor : Monitor & audit every action Claude Code takes on your computer.

A support library for Ronin. Like activesupport, but for hacking!

LLM speculative inference server for heterogeneous hardware & consumer GPUs