Lists (1)
Stars
Keep Claude Code's always-loaded context lean: deterministic memory guard-rail distiller + skills audit. Never deletes; dry-run default.
TokenSpeed is a speed-of-light LLM inference engine.
🔥 A Survey on AI Auto-Research
pool is Poolside’s coding agent that runs in your terminal or integrates with any ACP-compatible editor
Production-grade engineering skills for AI coding agents.
Pure shell + tmux + git event driven multi-harness orchestrator
A SOTA quantization algorithm for high-accuracy low-bit LLM inference, seamlessly optimized for CPU/XPU/CUDA, with multi-datatype support and full compatibility with vLLM, SGLang, and Transformers.
SQLite extension + bindings for Postgres NOTIFY/LISTEN semantics with durable queues, streams, pub/sub, and scheduler
Agentic search with ChromaDB and Context 1 model
Vibe-coded utilities for working with Nvidia's internal SASS architecture specification
Miles is an enterprise-facing reinforcement learning framework for LLM and VLM post-training, forked from and co-evolving with slime.
The theory of LLM wikis, running as one. A framework for agent-operated knowledge: typed, linked, review-gated markdown your agents execute.
Performance-optimized plugin packaging of GSD (Get Shit Done) for Claude Code. Based on open-gsd/get-shit-done-redux
Community-contributed instructions, agents, skills, and configurations to help you make the most of GitHub Copilot.
Fast LLM speculative inference server for consumer hardware.
SemaClaw is an open-source framework for general-purpose personal AI agents.
🪨 why use many token when few token do trick — Claude Code skill that cuts 65% of tokens by talking like caveman
A light-weight and powerful meta-prompting, context engineering and spec-driven development system for Claude Code by TÂCHES.
⚡ Native MLX Swift LLM inference server for Apple Silicon. OpenAI-compatible API, SSD streaming for 100B+ MoE models, TurboQuant KV cache compression, MACOS + iOS iPhone app.
Open-source local-first AI agent for desktop work. No account, no telemetry: use local models with Ollama/Rapid-MLX or bring your own provider key.