Trending Repos

Open-source end-to-end GEO implementation: status analysis, diagnosis, strategy, tickets, execution, verification

Self-hosted SIM/eSIM gateway for VoWiFi calling, SMS, cellular data and isolated regional egress

Self-hosted network monitor - hourly speed tests, LAN device counts via ARP scan, and sarcastic AI-generated reports delivered to Telegram or Discord.

A 股短线复盘看板：涨停池·连板梯队·龙虎榜·板块资金一屏看完，赚钱效应/晋级率/梯队断层/情绪周期等派生指标纯计算直出（不经过 AI），AI 只把数据串成能读的盘面研判。全本地运行，可用 Claude/Codex 订阅免 API key。| A-share short-term daily-review dashboard: derived sentiment metrics computed locally, AI only writes the narrative. No API key needed.

让智能体像人一样学习和工作

GEOHub: open, evidence-bounded GEO and SEO agent skills for AI Search, with research-grounded discovery, diagnosis, content, measurement, and one-line SEO planning.

Self-hosted AI workbench for knowledge, RAG, model providers, and safely governed user-built agents. Public Preview; production Agent Runtime remains gated. 自托管 AI 知识工作台与安全 Agent 平台。

Lobster0 — a self-hosted OpenClaw-inspired AI agent with persistent memory, governed autonomy, and self-evolution. 受 OpenClaw 启发的自托管 AI Agent「Lobster0」，具备持久记忆、受控自治与自我进化能力

Local-first cinematic visual archive for filmmakers — search stills by look, mood & technique; ingest video/URLs; moodboards. Own your frames (FilmGrab / Flim / Kive style, on your machine).

Local-first agent workspace for knowledge, tasks, documents and verifiable workflows · Локальная агентная система для знаний, задач и проверяемых процессов · t.me/romarayt

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

Community registry of LLM serving-path traps that produce confidently wrong measurements: templates, tool parsers, reasoning fields, quant kernel paths, CUDA toolchains, KV allocation, eval harnesses, versioning. Symptom-first, with the check that catches each.

Adaptable agent teams. Visually

Zerolinear — AI research lab and home of the Zeline open-source agentic AI framework.

A self-hosted, read-only IBKR trading decision, risk-control, and review system.

Self-hosted memory server for AI agents. Your brain is a git repo, served over MCP and REST.

Benchmark your local LLMs: speed, memory, and quality, in one command. TUI leaderboard for Ollama, LM Studio, llama.cpp, and vLLM.

Self-hosted job monitor with Telegram alerts｜追蹤公司官網職缺，符合條件就推送到 Telegram

Self-hosted, Docker-first OpenAI-compatible LLM gateway with provider routing, fallbacks, API keys, and usage controls.

跨 agent 的 skill 管理器 + skill 推广工具，欢迎fork和pr。管理你所有 skill ，创建技能自带讲解pdf或ppt。支持各种Agent。

Catch procedure errors before they become incidents. Train and evaluate small VLMs for temporal grounding in first-person factory video.

The acquisition pillar of Cartridge (by MoveWeight): request a ROM, ROMarr finds it via Prowlarr, grabs it, files it into RomM/Gaseous/Retrom. Unofficial.

飞影 feiying — 发个片名自动入库到飞牛/Emby:Telegram 多源发现 + AI 智能匹配 + 缓存流播 + 剧集自动追更,单 Docker 容器

Native PowerPoint (.pptx) generator tool for Open WebUI — JSON spec in, native charts/icons/layouts out. MIT.

Capable, auditable coding that runs fully offline on a 16 GB machine. A verification-first layer (hard test execution, symbolic checking, agentic repair) that takes a local 7B to parity with its 671B teacher on verifiable tasks. MIT, pre-registered, reproducible.

Optimized local serving engine for Kimi-Linear-48B: INT4 quantizer, fused decode kernels for a measured 3.18x, and an OpenAI-compatible server. Ships with k3, a bridge that detects the client per request so Claude Code, Codex, Cline, Aider and opencode all work unchanged.

让 AI 真的看见你发的视频：抽帧喂给模型，以及自建服务器上传视频的两个隐形坑

A self-hosted, private-first LLM companion and memory assistant.

An early prototype alternative to Microsoft/Windows Recall, built to run 100% locally with zero cloud communication. Captures screen snapshots, extracts text via WinRT OCR, and indexes embeddings into SQLite. Query your history conversationally via your local LLM (LM Studio). Under active development.

Open-source, self-hosted LangGraph Agent Server. Production Postgres + Redis. Same SDK. Same Studio. Zero code changes.

自托管的安全视频解析与授权媒体下载工具，基于 Next.js、FastAPI、yt-dlp、FFmpeg、Redis 和 RQ。Self-hosted secure video inspector and authorized media downloader.

Truthful scheduler-step and streaming visualization for local LLM inference

Self-hosted web file manager backed by your private Telegram channel, with Server Files for VPS and SFTP access.

Production-grade SEO + GEO content factory for Claude Code — research, write, fact-check, optimize, publish to WordPress, monitor. Battle-tested by Loamwright 沃匠 SEO agency.

Self-hosted multilingual (English + 5 Indic) phone voice agent template - LiveKit Agents + Sarvam STT/TTS + OpenAI + Vobiz SIP. ~700ms-1.2s latency, under Rs 5.5/min. Demo: a real-estate site-visit booking agent.

An AI portfolio copilot that unifies multi-account, multi-market holdings for smarter analysis, risk diagnosis, and investment review.

Async Python CLI & library for FREE web AI services - No Login, No Api Key, No Browser Required

🦞 Headless real-time Python voice assistant powered by ElevenLabs, Gemini, Mem0, MCP tools, and local wake-word detection.

Session-scoped GPT-5.6 agent routing for Claude Code

A self-hosted DuckDB-as-a-Service server with REST API and Web UI.

Self-hosted, OpenAI-compatible speech server: faster-whisper STT and Piper/Kokoro TTS. A local voice backend for Open WebUI, CLIs, and custom agents.

Local-first AI that can actually use your computer.

AI brain for your Paperless-ngx archive — chat with your documents via local (Ollama) or cloud LLMs, with agentic tool use, semantic search, and long-term memory.

Admin-approved, audited SQL gateway — Slack + web IDE, tier-matched creds (RO/RW/DDL), PII masking, full audit. PostgreSQL & SQL Server.

Self-hosted, multi-user agent platform for teams: skills, knowledge and API contracts live in the database and grow at runtime. Typed Python core (Protocol ports) you can embed in your own stack.

Export UniFi Network topology to SVG, PDF, PNG, HTML, DOT, Mermaid, JSON, and editable draw.io diagrams.

Annona — the sovereign execution kernel for AI agents. Decides where each step runs, enforces it, and records it.

Open-source asset inventory — turn scattered, contradictory tool data into one reliable inventory, then chat with it using the LLM of your choice. Tagging, posture rules, and alerting built in.

Catches silent CPU fallback and mislabeled tok/s in local LLMs. llama.cpp and ollama, one file, no deps.

Give your CMaNGOS/playerbots AI companions real personalities with a local LLM — a tiny, dependency-free Python shim over Ollama. Per-bot personas, memory, and bot-to-bot banter, fully offline.

주제별 핵심 개념 이해부터 직접 실행까지 다루는 가이드·튜토리얼·실습 모음

Raspberry Pi based drink inventory system with barcode scanning, stock tracking, statistics and a web interface.

Hikari's curated English measured notes on local LLM systems (serving, specdec, training, steering).

An institutional-intelligence operating system: a council of 14 agents that remembers every deliberation, grades its own predictions against reality, and revises what it believes. Runs on your machine — Ollama, llama.cpp, or any cloud API.

Qwen 3.8 4B (Distilled): Run The Open Source King Locally! - Offline prompt engineering engine.

Local LLM speed measurements from a 4x RTX 3090 rig, with exact launch commands and per-run provenance.

Get tuya local keys and complete device information for every tuya device via QR code login (no tuya Developer account required). Features a Web UI for Home Assistant and Docker, plus a CLI.

Plug-and-play, dependency-free web command center for DGX Spark fleets (and any GPU boxes): live per-node GPU/CPU/mem/temps/power, optional RouterOS switch monitoring, and per-model decode/prefill/TTFT/KV cards. Point it at your own hardware.

GRID v2 — the local-first AI agent that RECONS THE WORLD. 18 capabilities & 72+ tools in one terminal: OSINT, network recon, SDR/radio, satellite, IoT, computer vision, offline CTF sandbox, game worlds (GRAD open-world + MiniGrid) — with a human-like 4-tier memory. Hermes writes code, OpenClaw reads DMs, GRID recons.

Laguna S 2.1 (NVFP4) on DGX Spark GB10: 20-cell tuning sweep, 12h soak (3,099 turns, raw logs), thinking-gate dose-response study, Qwen 35B head-to-head, quant-floor reference. Losers published too.

Discover trusted local AI models that fit your Windows PC — bilingual desktop app and CLI.

Measure whether ChatGPT, Claude, Gemini and Perplexity recommend your brand - and fix the reasons they don't. A free Agent Skill for AEO/GEO audits. No account, no database, Python stdlib only.

Local, offline, LeetCode-style interview prep for data science & ML — SQL/Python/ML/system design with a local-LLM mentor (Ollama). Zero cost, zero cloud.

A Python SDK for the iM.Master robots, with an optional local-LLM + vision addon.

Pure-Python AI coding agent powered by DeepSeek V4 - Subagent-First architecture · Context is Currency · Terminal-native - Can connect Ollama models for Vision support - Uses the native web search built into DS v4 API and is built around the DeepSeek caching mechanism