Medieval village economy simulation — up to 1000 LLM agents across multiple villages, driven by supply chains, hunger, tool degradation and market pressure.

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.

TimesFM (Time Series Foundation Model) is a pretrained time-series foundation model developed by Google Research for time-series forecasting.

A GPU-rendered terminal emulator with inline 3D graphics 🐀🧀

the formatter multiplexer [maintainers=@zimbatm,@brianmcgee]

Edit your flake inputs with ease.

A script to achieve automatically following all flake inputs for Nix

agent multiplexer that lives in your terminal.

A terminal workspace with batteries included

Yazi and Zellij with smart defaults & awesome plugins give helix/nvim a powerful yazi sidebar, git integrations, a configurable popup system (lazygit, a config ui, etc), zoxide integrations, zjstat…

Wraps any OpenAI API interface as Responses with MCPs support so it supports Codex. Adding any missing stateful features. Ollama and Vllm compliant.

Fork of teabranch/open-responses-server. OpenAI Responses API proxy for self-hosted LLMs (Codex CLI compatible)

Proxy intercepting and fixing malformed tool call from Qwen 3.x models served by vLLM. Support streaming responses.

htop for your whole fleet — with nothing installed on the remote

A lightweight, single-binary LLM inference engine built in Rust. It offers low-latency token streaming, continuous batching, and memory-efficient caching via an OpenAI-compatible API

AI-Powered, Non-Intrusive Terminal Assistant

Render markdown on the CLI, with pizzazz! 💅🏻

🌻 A Vim alignment plugin

Community recipes for serving LLMs on RTX 3090/4090/5090 CUDA gpus. Multi-engine (vLLM, llama.cpp, ik_llama) and model-agnostic. Currently shipping Qwen3.6-27B Qwen3.6 35B Gemma 4 26B Gemma 4 31B c…

MiniJinja is a powerful but minimal dependency template engine for Rust compatible with Jinja/Jinja2

A Haskell implementation of the Jinja template language.

c2hs is a pre-processor for Haskell FFI bindings to C libraries

Get your repo ready for AI.

Agent Package Manager

My near-autonomous workflow for coding agents

Convert Styled Layer Descriptor (SLD) files to mapbox layer json

Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI.

Windows-native Nix evaluator - Haskell for logic, C99 for data. Parser, lazy evaluator, content-addressed store, builder, binary substituter.