Real-time system audio translation for macOS — translate any audio (YouTube, podcasts, meetings) live on screen with OpenAI or Google Gemini. On-device speech recognition; optional low-latency real…

A framework for generating realistic LLM serving workloads

Code search MCP for Claude Code. Make entire codebase the context for any coding agent.

A reactive notebook for Python — run reproducible experiments, query with SQL, execute as a script, deploy as an app, and version with git. Stored as pure Python. All in a modern, AI-native editor.

GitHub Copilot extension for JupyterLab

🪐 🔧 Model Context Protocol (MCP) Server for Jupyter.

A CLI tool to switch and manage Codex accounts

Implementing DeepSeek R1's GRPO algorithm from scratch

An Efficient and User-Friendly Scaling Library for Reinforcement Learning with Large Language Models

DualMap: Enabling Both Cache Affinity and Load Balancing for Distributed LLM Serving

Official implementation of ARDY: Autoregressive Diffusion with Hybrid Representation for Interactive Human Motion Generation (SIGGRAPH 2026).

A high-performance and light-weight router for vLLM large scale deployment

FlashKDA: high-performance Kimi Delta Attention kernels

Compiler and runtime implementation for PIM device.

A retargetable MLIR-based machine learning compiler and runtime toolkit.

This repository is to provide graph mode execution with GPU+PIM in runtime.

This repository relates to SYCL support for PIM

Anthropic's original performance take-home, now open for you to try!

Build a compiler to solve Anthropic's interview challenge.

SW Library for Samsung PNM (including functional simulator)

🛰️ Track token usage across AI coding agents from your terminal. 🏅 Global leaderboard with trillions of tokens tracked.

An open toolkit and public dataset hub for collecting, sanitizing, analyzing, and visualizing coding agent traces.

LLM KV cache compression made easy

vLLM wall-clock emulation pluging to allow running the real vllm serve and benchmark without gpu with simple mocking