Havuz

🌊 34 free curated AI models (up to 1M context) for Oh My Pi & Pi. Thin by design — model list, relay proxy, tool-call diagnostics — add your

A tiny Claude Code skill that keeps your prompt cache warm during idle sessions, so your next message reads from cache instead of paying ful

Context management proxy for AI coding agents: verbatim archive, on-demand recall, and saved engine state on supported backends.

ASD-STE100 Simplified Technical English as a plugin for Claude Code, Codex & coding agents — clearer, unambiguous LLM output

Open-source privacy firewall for AI agents: masks personal data in text, tool calls and images before it reaches the LLM, and restores it in

A decision cache for TypeSafe Jev-class models — memoize decisions so repeats are free, deterministic, and shareable. One 2 MB binary.

Console is a private, native Android app for your self-hosted Hermes Agent: chat that shows the agent's work, voice, bots and rooms, cron, t

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

Run large language models locally on Intel Macs with AMD GPUs - native macOS app with Metal acceleration

🧬 Symbio（共生）- AI Infra 级多 Agent 协同框架 | 33 Killer Features: 动态 DAG, 本体化记忆, 防过早完成, 数据飞轮, 神经符号安全, OpenTelemetry 可观测, HITL 人类介入, MCP 原生支持, 语义缓存,

OpenCode Mobile — open-source Android client for the OpenCode AI coding agent. Run AI coding sessions from your phone against your self-host

Turkish Native, AI ve yerelleştirme metinlerindeki çeviri kokan Türkçeyi düzeltmek için agent'lara yönelik bir yazım becerisidir.

Neden yükseliyor · Türkçe çeviri kokusu gibi belirgin bir soruna odaklanıyor; kurulum yönergeleri, örnekleri ve regression case'leri kullanımını somutlaştırıyor. 16 günde 88 yıldız ve 32 commit erken ilgi ile yoğun geliştirmeye işaret ediyor; ancak commit'lerin tek yazardan gelmesi bağımsız katkı veya doğrulama kanıtı sunmuyor.

AgentScope Java, Java ile uzun süre çalışan ve dağıtık ortamlara taşınabilen agent uygulamaları geliştirmek için bir framework sunuyor.

Neden yükseliyor · 13 günlük yaşına rağmen 108 yıldıza ve son 28 günde 12 commite ulaşmış; Java tarafında üretim odaklı agent altyapısı arayanların ilgisini çekmiş olabilir. Ancak yıldız ve README özellikleri benimseme ya da üretim başarısının kanıtı değil; farklı yazar sayısının 1 olması katkı tabanının şimdilik dar olduğunu gösteriyor.

标书匠 BidCraft · 对话式标书编制系统：AI 智能编制技术标，支持招标文件解析、标段选择、标书编写、知识库提炼与学习对话式改稿。LLM-Wiki 路线，基于LangGraph实现。Open-source AI agent for construction tender

Two AI VTubers co-hosting one live show — the realtime engine & orchestration. Open-core, battle-tested by a daily stream.

OpenCode Android — unofficial client for opencode. The real coding agent runs natively on your phone: no Termux, no PC, no root. One arm64 A

TestOPS一站式AI运维平台 · By 一鸣（公众号：一鸣AI编程）

Custom APRL machine-learning algorithm + a curated ML knowledge base of 923 papers, lectures, and explainers. NumPy classifier/regressor, te

Serve your models fast on your own infra in 3 steps.

The one place for the skills, keys and memory of every agent you use. Formerly OMNI.

Scaler 2 Music Theory Tool 2026: Direct DAW Integration for Windows 10 and 11

Fast ZON to JSON Converter for Modern Apps 2026

OpenJev: open-source, local alternative to Jev. System-One decision models: typed questions in, calibrated probabilities out; general decisi

Connect Luau Games to 100+ DataGrout Logic Cells and Batteries 2026

The fastest inference engine for Qwen3.8-27B on the NVIDIA RTX 5090: up to 500 tokens/s for one agent and up to 2,000 tokens/s for many, con

开源私域 AI 营销 SCRM 系统，Martech 获客引流利器， 智能AI客服 、销售 ReAct 自主 AI 智能体 全渠道聚合（抖音/小红书/快手/微信/WhatsApp/TikTok/Telegram）统一收件箱 独家 GEO 模块（生成式大模型引擎优化/AI 搜索占位

卡网大全：一站式 AI 大模型账号购买导航。聚合 AI 中转站、卡网、官方订阅比价，还有模型排行、使用指南、实用工具，帮你少踩坑，少花钱。

Lightweight AI-native workflow builder for individuals and small teams — describe your idea in natural language, get a runnable workflow on

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

Open-source alternative to TypeSafe's Jev: a System One style model layer that gives typed, calibrated decisions from any open-weights LLM i

Open-source, self-hostable alternative to Claude Tag — a Slack-style workspace where your team and its AI agents (Claude Code, Codex, GitHub

Distill a repeated Jev classification task into a local model, on the fly — same answers, your hardware.

👾 下一代透明智能体架构 | Next-Gen Transparent Agent Architecture 🔍 全行为审计 | 🛡️ 两段式安全调用 | 🧠 双水位记忆 | ⏰ 心跳任务 📊 P0 级事故率降低 80% | 兼容 OpenClaw + Claude Code 技

Drop-in prompt compression for production LLM apps. Cut your token bill 40-60% without changing your code. Python SDK, LLMLingua-2, MIT.

Turn a model gateway into a standard OpenAI-compatible API with a built-in web console. Zero deps (Node.js built-ins only). 把华为云果办 OfficeAce

A local-first workspace for documents, tasks and meetings on macOS — nested spaces, typed properties, relations and saved views, with transc

Persona agents with explicit memory, retrieval, adversarial smoke tests and a practical post-training roadmap.

Free, local, open-source LLM cost analyzer - see where your LLM bill leaks, on your machine.

A persistent, reviewed multi-agent runtime for long-horizon research and engineering.

Project Tapestry aims to give every nation and participant frontier AI they can call their own — uniting a global consortium to train a shar

Give the agent "this" and he gets a real PC.

Turn any local language model into a fast decision API with typed, probabilistic answers

从 MiniMind 源码读起,再延伸到现代大模型技术体系的中文学习笔记。主线逐行精读预训练 / SFT / DPO / PPO / GRPO 与训练机制;附录 17 篇进阶卷覆盖量化、投机解码、RLHF 全景、模型代际史等 MiniMind 没涉及、但进阶绕不开的主题。

Open-source AI chatbot & site search for your docs. One <script> tag, streaming answers with sources. 100% Cloudflare-native (Workers AI + V

不教 PUA，只教真心。Win your crush before it's too late. Inspired by colleague-skill（同事skill）

Native Android AI agent: use your own API keys (OpenAI/Anthropic/Gemini), opencode-style modes, device tools, Shizuku shell, screen analysis

风插 2.0 — DSH 推理语域锚定预设：两个 agent preset（风插标准/风插 PTC）在 runtime-context 快照头尾各注入一次形式化约束，使推理以第一人称复数展开。定性=硬指令约束（非吸引子）；未证明能力提升，未证明无负面作用。Apache-2.0

PyTorch 기반 한국어 LLM 구현 및 학습 프로젝트

Turn scattered notes, docs and transcripts into a queryable Markdown wiki — an LLM knowledge-base compiler with MCP access, no embeddings, s

Universal, model-agnostic operating harness for AI agents (Claude, Codex, Gemini, …) — a lean core + work-type profiles assembled by one set