Havuz

Official Code for OpenVLHarness: Real-World Visual Agentic Tool‑Use with Multimodal Memory

Make any NPC in any game talk back with Fish Audio: skills, a self-hosted voice server and tools for AI coding agents, with a full GTA V exa

一个会主动发现任务的智能体

咖啡伴侣：《活侠传》全能Mod - AI陪玩、剧情回顾、分支自选、修改面板、存档优化 / 咖啡伴侶：《活俠傳》全能Mod - AI陪玩、劇情回顧、分支自選、修改面板、存檔優化 / Coffee Companion: All-in-One Mod for Legend of Mor

Open-source platform for AI chat agents that answer from your own knowledge base. Hybrid RAG search in PostgreSQL, MCP and custom tools, gua

AI workers that turn GitHub issues into verified bug-fix pull requests: reproduce, fix with a regression test, prove it, open the PR. Runs l

A benchmark for AI agents as teammates. Paper: https://arxiv.org/abs/2611.02633

One native app for every AI coding agent. Rust + GPUI.

🧠 Persistent local-first memory for AI coding agents — MCP server + CLI for Claude Code & Cursor. SQLite FTS5, secret redaction, project iso

An easier way to use AI on your own computer. A free, open-source alternative to Claude and ChatGPT that runs open-weight models like Gemma

Replay your coding agent sessions as a map and a timeline

Skill for AI agents that draws publication-quality scientific schematic figures as vector code

DecisionPivot: a diagnostic benchmark for decision models in LLM systems. Contrast pairs that differ in one decisive detail, generators with

A language model trained from scratch and running offline on an ESP32-S3 with a touchscreen chat interface.

Qwen3.8-Flash-Next (180B MoE) on one 12-16 GB NVIDIA GPU, Linux or Windows. Own engine, own compression.

Lev: Thinking Fast and Deep with Looped Decision Models. One checkpoint, multiple depths: System 1.5 decision models that think in loops, no

AI-native, mixed-reality, self-evolving open world engine. 11-layer architecture, 5 platforms (PC / Mobile / Web / AR / BCI-reserved).

DSH 防退化套件｜不止打断，还会清创：把退化的思考块从模型可见历史里折叠掉，下一轮不再回传。复读分两型——A 型同一句死磕、B 型短句轮换（最难抓，实测英文 8 块全漏）。两个插件一套闭环，缺一个都会漏。

Open-source AI studio for short dramas, short videos and product ads: one-line idea → script, cast, storyboard, video, voice, edit. Self-hos

Train JEV-style System 1 decision models on Qwen3.5 and Qwen3.8-27B: 24-slot decision head, KL distillation, calibration and serving. 在 Qwen

Experience-Driven Black-Box Cost Control for Coding Agents

AI 跨境电商运营智能体工具箱 · 把运营 SOP 拆成 7 个可复用的 AI Agent 工作流，含 Prompt、模板与真实案例

Scout: the open-source browser agent, by Outcrawl. Mirrored automatically from Outcrawl's own repository.

SpeakMore：口述成文，划词即译。开源、免费的语音输入与翻译工具，支持 Mac 与 Windows：语音输入、语音翻译、划词翻译。兼容 Gemini、OpenAI、通义千问、DeepSeek、Kimi、智谱 GLM、MiniMax 及任意 OpenAI 兼容服务，语音输入亦

Build autonomous AI agents that interact with blockchains — in minutes, not months.

OpenAI-compatible bridge for OpenCode free-tier models — no account required

Local decision engine with typed, calibrated outputs (Noul/Choice/Score): cheap→strong cascade, ECE/Brier calibration, SystemOne protocol in

🔒 Your private AI assistant, 100% local on Windows: chat, vision, documents (RAG), web search, images, videos with sound, voice, code projec

The open source AI coding agent, in Rust. CLI, TUI and HTTP server for local AI agents — no Docker required.

确定性取证 × 大模型定性 × Agent 反证复核的恶意代码审计智能体 · Deterministic forensics × LLM verdict × agent falsification for malicious-code audits

Native AI assistant for macOS, built with SwiftUI and llama.cpp. Run local LLMs offline, work with PDFs, and create custom AI assistants. Op

Cromford: the open-source software factory. Tickets in, reviewed, merged and production-checked changes out.

三菱 PLC 编程与工程助手。 MCP server and AI workbench for Mitsubishi PLC programming with GX Works2. Generate Ladder and Structured Text, edit native

Carefully human hand written llm jailbreaks for the latest models that are available on the market. GPT-6 Astra, GPT-6.1 SoL, Opus 4.7, Opus

AI 模拟面试官：每条扣分都强制引用你回答的逐字原文（机器校验，编造引用会被重试/丢弃），四维度加权由确定性代码计算而非模型现编，评分体系设计为可经真人盲评校准 · FastAPI + Vue3 + LangGraph

A small chat model trained from scratch: tokenizer → pretraining → SFT → RL → multimodal. 0.78B params, CORE 0.248, tool calling, vLLM. Weig

Illustrated comic-style cheat sheets for ML, LLMs, AI safety, LeetCode and Git, plus a SwiftUI study app (Panels) to read them and quiz your

A design system for AI applications. Material, physical, tactile, technical.

A multi-model Muse: personal AI voice assistant on ESP32 - wake word or knob, any OpenAI-compatible brain (STT/chat/TTS mix-and-match), stre

Local-first AI agent for Chrome: browser automation, web research, and data extraction with your own model.

Say "¥35, not spicy, no ice" → one answer + a copyable order card. Official calculate-price only. Zero-dependency MCP skill. 麦当劳点餐 Skill.

An end-to-end framework for Agentic Reinforcement Learning.

120 patterns that make writing read like AI slop, with an agent skill for Claude Code and Codex.

A dishwasher-loading game on a physical simulation of one real machine, built to benchmark language models and playable in a browser. The ga

Never hit rate limits again. One OpenAI-compatible endpoint for 60+ providers & 1,200+ models — smart auto-fallback, prompt compression, 60-

Run the official Claude Code natively on Android — no root, no proot. Chat GUI, terminal, toolchain and APK builds, all on your phone.

Which AI coding model, at which effort, on which plan? Recommended efforts, a leaderboard at those efforts and plan advice from modelplan.ai

Text-to-CAD tools in Elixir with build123d, OpenSCAD, replicad, and smlibpy.

Self-hosted AI pull request review bot on Pi Durable. Comment /dureview to get P0–P3 inline findings and a check run, using your own ChatGPT

MCP server that lets AI agents operate Moodle: courses, assignments, forums, grades and files-as-markdown.