N.E.K.O and NekoACM
The two tools are **homonymous but unrelated projects**, sharing the name "N.E.K.O." but addressing entirely different problem domains: one is a multi-modal AI companion, and the other is an AI toolkit for competitive programming.
About N.E.K.O
Project-N-E-K-O/N.E.K.O
N.E.K.O. — A proactive, multi-modal AI companion featuring 24/7 ambient awareness, agent capability and an embodied emotional engine.
Built on a modular architecture supporting multiple LLM providers (Qwen, OpenAI, GLM, StepFun) with pluggable emotion and memory systems, N.E.K.O. synchronizes conversational context across desktop, mobile, and metaverse applications through persistent state management. The project provides zero-config Steam deployment alongside Docker infrastructure for self-hosted Linux installations, with extensible UGC integration via Steam Workshop for community-contributed models, voices, and personality packs.
About NekoACM
HEX9CF/NekoACM
🐱🐾NekoACM: Neural-network Engine Kit of ACM-ICPC. 基于人工智能大模型的 ACM-ICPC 神经网络引擎工具包，提供自动出题、生成测试数据和题解代码、题目格式解析、题目国际化翻译、虚拟评测机、AI 编程助手等功能。可以作为模块集成到 Online Judge 系统中。
Built on OpenAI-compatible LLM APIs, NekoACM operates in three modes—CLI, interactive chat, and HTTP server (port 14515)—enabling both standalone use and modular integration into Online Judge systems via REST endpoints. The platform leverages LLM context-awareness to handle algorithmic problem translation with domain-specific terminology accuracy, and provides an AI-powered code sandbox that predicts execution results without actual compilation. JSON-based problem representation enables seamless format conversion and multi-language code generation across C/C++/Java/Python and other languages.
Scores updated daily from GitHub, PyPI, and npm data. How scores work