Lists (1)
Stars
My MSc experiments benchmarking GPTQ quantization on Llama 3.2 1B
Handheld RISC-V Mini Game Console with OLED Display
KVarN is a native vLLM KV-cache quantization backend for your agents: 3-5x more context, throughput above FP16, and FP16-level accuracy. Calibration-free, one flag.
Header only library for writing build recipes in C.
Any model. Any hardware. Zero compromise. Built with @ziglang / @openxla / MLIR / @bazelbuild
A boilerplate project for TheFloW's so-loader to save time on common project setup for Android to Vita ports
A Simple Streamlit Demo for Serving Supertonic-3 Using Streamlit
TinyStories-260K running locally on a stock Game Boy Color.
Inference Llama 2 in several files of pure C but on a C64
A Cargo-inspired build tool and package manager for compiled languages that target GCC or Clang. Freight handles C, C++, Fortran, CUDA, HIP, OpenCL, ISPC, and assembly — with a single declarative f…
Provide free GPT-3.5 API service by reverse engineering the login-free ChatGPT website.
DeepSeek 4 Flash and PRO local inference engine for Metal, CUDA and ROCm
A self-paced course to learn Rust, one exercise at a time.
A local MCP server and CLI tool for scraping Italian real estate listings. Built with a plugin architecture — adding a new site means dropping in one config file.
Artifact mirror for the Memory of Alessa project
Vane is an AI-powered answering engine.
General plug-and-play inference library for Recursive Language Models (RLMs), supporting various sandboxes.
A proof-of-concept project that showcases the potential for using small, locally trainable LLMs to create next-generation documentation tools.
Khronos version of the Vulkan Tutorial as parts of the Vulkan documentation project
Agent harness to make your slop code well-engineered and beautiful.