Self-hosted API for your RL Infrastructure

These are commands I use with agents, mostly Claude

A benchmark built to evaluate and improve agent capabilities for supporting legal work.

AI agents running research on single-GPU nanochat training automatically

GPU-CR: GPU Checkpoint & Restore

code repo for GCR [FAST'26]

Ludic – an LLM-RL library for the era of experience

Training API and CLI

A benchmark to evaluate LLMs on Kubernetes tasks

Post-training with Tinker

MoE training for Me and You and maybe other people

Streamline on-policy/off-policy distillation workflows in a few lines of code

Data recipes and robust infrastructure for training AI agents

CUDA checkpoint and restore utility

A modern Anki custom scheduling based on Free Spaced Repetition Scheduler algorithm

Plugin index for https://github.com/kubernetes-sigs/krew. This repo is for plugin maintainers.

LLM-generated real-time commentary for Pong

This repository contains the website implementation and content for the GKE AI Labs site. GKE AI Labs is a collection of examples, best-practices, and prebuilt solutions to help build, deploy, and …

Run Claude Code on OpenAI models

Readline is a pure go(golang) implementation for GNU-Readline kind library

the TypeScript AI agent framework