Havuz

Privacy-first LLM proxy and AI gateway - load balancing, multi-provider routing, API key management, usage tracking, rate limiting. Self-hos

211 free visual cards about large language models, from tokenization to deployment. Full collection: 376 cards.

Flash weight streaming for MLX: run massive models larger than your RAM on Apple Silicon.

带 AI 精读大型开源仓库的方法论：四阶段流程、可复用模板、28 条踩坑清单，核心是让每个技术论断都可回溯到源码具体行

为 DeepSeek 前缀缓存定制的终端 Code Agent(纯 Go),缓存命中率 95-99%,命中输入定价为未命中的 1/30。A terminal coding agent optimized for DeepSeek prefix caching — 95-99% c

Who is really behind that API? Fingerprint any OpenAI-compatible endpoint in the browser: 9 infrastructure probes, side-by-side comparison,

The first open-source repository for Scientific Machine Learning (PINNs, PIELM, RFM, Neural Operators) via Vibe Coding. JAX-GPU implementati

Open-source AI business orchestration platform for running governed AI agents across ERP, CRM, Microsoft 365 and other business systems.

Open-source AI API Gateway | Serverless & VPS Self-hosted | Vercel, Cloudflare & Local | Multi-Provider, Smart Routing, Key Rotation & Failo

轻量、自托管的多入口 AI Agent 机器人｜QQ 官方 / OneBot 11 / NapCat / 微信｜Rust｜Memory / RAG / Todo / Tool Calling

🐙 盯着 Claude Code 的桌面宠物：随 agent 状态变表情、弹消息气泡、一键授权，并统计 token 用量与花费。本地优先、MIT。

MemPrivacy is a privacy-preserving personalized memory management framework for edge-cloud agents.

Simulate AI Personas in Social Feeds with Real-Time Chat Bots 2026

Anaxa 是一个面向科研工作流的开源智能体系统。它不是单纯的聊天机器人，也不是无人监管的自动发论文机器，而是把文献检索、证据审计、实验执行、论文写作、同行评审式检查和最终产物打包放进同一个可追踪的研究生命周期中。

MilkyAi at bilibili🥛| Video-to-Notes Email Service with RAG Customer Support & 38K Followers/Users on Bilibili | Python, PostgreSQL, RAG, A

原生AI股票研究平台，支持多智能体分析、AI Agent 协作与实时辩论对抗，可配置的大语言模型（LLM）工作流、多源市场数据及报告导出功能。

A cognition aware database engine for AI agent memory. Purpose built in Rust with WAL, HNSW, knowledge graphs, and speculative context pre a

Swift SDK for running chat, vision and speech models on iPhone and Mac with Apple's Core AI. Model download and caching, FoundationModels in

One window for all your AI coding agents. Run Claude, Codex, OpenCode, Gemini, Antigravity, Cursor, and Copilot side-by-side. Terminal and c

Self-hosted AI chat platform with multi-step tool calls, persistent Python sandbox, RAG, and Deep Research. Supports Claude, GPT, Gemini and

A tiny neural-network framework in pure Go with AVX2 SIMD kernels (GOEXPERIMENT=simd)

Self-hosted AI workspace with agents, skills, and tools (Gmail, Calendar) that runs entirely on your own provider API keys (BYOK). Bring you

Translate AI garbage to human language. A Pi package for plain-language rewrites.

An "LLM wiki" upgraded to a real database — typed entities, graph relations, HTTP API, and a built-in natural-language agent.

One command gives AI agents instant codebase context. ~250 tokens replaces 50,000+ tokens of exploration. Auto-configures Claude Code, Curso

Offline-first Python AI agent that runs a tiny research business: quotes each job against its own costs, collects via Stripe, fulfils with N

Qwen-Planner-Agent: A Closed-Loop AI-for-AI Framework for Real-World Mobile Planner Agents

Local LLM cost-tracking proxy for OpenAI, Anthropic, Gemini, and pinned OpenRouter calls with token usage, failure, and billing-integrity re

Token Bank — the local LLM gateway that sits between your AI agents and every provider. Know where tokens go · Spend less with smart routin

The original AI Chief of Staff built on Claude Code and Obsidian. ADHD prosthetic, open to anyone. Superseded by chief-of-staff-2.

On-device AI meeting assistant — real-time transcription & LLM insights, 100% offline

Policy guardrails for coding agents (Claude Code, Codex, Cursor) — every tool call is checked locally, before it runs.

The AI-friendly C23 compiler for security research, built on LLVM [WIP]

The open-core AI workbench — notebooks, agents, RAG, voice, and images across any model: OpenAI, Anthropic, Google, xAI, or local via Ollama

Travel MCP server + CLI for AI assistants — flights, hotels, trains, cars, ferries, price alerts. No API keys, single Go binary. 1 smart MCP

Pie – free, open-source AI browser agent for Chrome. Reads pages & PDFs, fills forms, extracts data, automates tabs, hands off to Claude Cod

Composable Agent Skills (Claude + OpenAI Codex) for taking an idea from fuzzy → validated → sequenced build → shipped — a manual tier (ideat

Recursive Self-Improvement for AI agents and models, helping your agents self-improve with every run

🧭 给 Claude Code / Cursor / Codex 用的『架构副驾』skill —— 开新项目时用持续深度提问引导你在写代码前想清楚架构(产出架构图 / ADR / 演进路线)。知识源自 awesome-architecture。中英双语。

Convert CodeBuddy/WorkBuddy (Tencent coding assistant) subscription into an OpenAI-compatible API. Use CodeBuddy GLM-5.2/Kimi/DeepSeek model

Nuphus — 本地优先的 AI Agent：真实桌面执行力 + 手机第二块屏幕。Local-first AI agent with real desktop execution and dual-device real-time sync.

SDKs and reference implementations for the Synadia Agent Protocol for NATS

EurekAgent: an autonomous research system for metric-driven tasks, built with Claude Code. Define the problem and metric. Get breakthrough r

Fine-grained robot control with Jev, physics previews, and configurable LIBERO tasks.

A finite ontology of 115 recurring software elements and a composition framework for LLM-based coding agents

First open-source implementation of Google TurboQuant (ICLR 2026) -- near-optimal KV cache compression for LLM inference. 5x compression wit

Open-source self-hosted home AI inference platform for AMD Strix Halo — multi-backend slots, OpenAI-compatible gateway, Vue 3 + FastAPI + sy

Benchmark self-evolving Agent upon realistic large-scale file workspaces

A compounding LLM-wiki, best fit as a personal second brain that organizes and updates itself as it grows, zero upkeep.

Local-first CLI for coding-agent context handoff. Transfer and resume sessions across Claude Code, Codex, Cursor, Cline, Kimi, Antigravity,