Stars
DeepSpec: a full-stack codebase for training and evaluating speculative decoding algorithms
claude code install and update on windows OS
Qwen-AgentWorld: Language World Models for General Agents
MCore-Bridge: Providing Megatron-Core model definitions for state-of-the-art large models and making Megatron training as simple as Transformers — with support for 300+ large language models (Qwen3…
FrontierSWE is an ultra long-horizon coding agent benchmark that tests implementation, performance eng and ML research
Darwin Gödel Machine: Open-Ended Evolution of Self-Improving Agents
Skills for Real Engineers. Straight from my .claude directory.
MobileGym: A Verifiable and Highly Parallel Simulation Platform for Mobile GUI Agent Research · 浏览器里运行的安卓模拟器 · Browser-hosted Android Simulator · Verifiable Evaluation · Scalable Online RL Training
official implementation of "Claw-Anything: Benchmarking Always-On Personal Assistants with Broader Access to the User's Digital World"
Advances and Frontiers of LLM-based Issue Resolution in Software Engineering A Comprehensive Survey
"OpenHarness: Open Agent Harness with a Built-in Personal Agent--Ohmo!"
Open-source Environment toolkit of claw-like agents, support task/harness generation and evaluation
Build, Evaluate, and Deploy GUI Agents — online RL training, standardized benchmarks, and real-device deployment in one framework.
Google Workspace CLI — one command-line tool for Drive, Gmail, Calendar, Sheets, Docs, Chat, Admin, and more. Dynamically built from Google Discovery Service. Includes AI agent skills.
Run agents like Hermes and OpenClaw more securely inside NVIDIA OpenShell with managed inference
The agent that grows with you
Claw-Eval is an evaluation harness for evaluating LLM as agents. All tasks verified by humans.
An in-the-wild benchmark for AI agents in the OpenClaw Environment.
很多镜像都在国外。比如 gcr 。国内下载很慢，需要加速。致力于提供连接全世界的稳定可靠安全的容器镜像服务。
Curated list of AutoResearch use cases with optimization traces and open source implementations
Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.
CRS-自建Claude Code镜像，一站式开源中转服务，让 Claude、OpenAI、Gemini、Droid 订阅统一接入，支持拼车共享，更高效分摊成本，原生工具无缝使用。
An Efficient and User-Friendly Scaling Library for Reinforcement Learning with Large Language Models
From Early Internet Design Patterns to AI Agent Implementation — A Deep Dive into Claude Code for Developers
"CLI-Anything: Making ALL Software Agent-Native" -- CLI-Hub: https://clianything.cc/