Run 35B MoE models in ~2.5 GB of RAM. Weights stream from SSD; a trained router predicts loads a step ahead. Mac, iPhone, Android, Windows supported
AirLLM 2.8T LLM inference with single 4GB GPU
🕷️ A Web Scraping framework that handles everything from a single request to a full-scale crawl! Don't be shy, join here: https://discord.gg/EMgGbDceNQ and follow here for daily tips and tricks: https://x.com/Scrapling_dev
红果短剧|Windows 10/11 桌面版 + Android TV 测试版。真人剧、漫剧、AI 剧;电视 APK 4.08 MB,Android 5.0+ / ARM64,支持遥控、续播与换一部;免费使用,由渠道有数独立维护。
红果短剧电脑版 · 不用安卓模拟器,直接在 Windows / macOS 上刷红果短剧 · 国内下载: hongguo.235698.xyz
TokenSpeed is a speed-of-light LLM inference engine.
一个先接住情绪、再分析关系并给出可执行策略的 Codex 恋爱军师,内置心理、法律、社会、人文、哲学、婚姻家庭与性学知识库,支持多元关系。
🚁🚀 2026机场推荐 | 熊猫VPN(PandaVPNPro)已确定跑路!快连VPN ,小牛加速器(小牛VPN)体验不佳且价格贵。低价机场,便宜机场,平价机场,性价比机场,高速机场,稳定机场,一元机场,翻墙机场,付费机场,收费机场,冷门机场,优质机场,廉价机场。机场评测,中国翻墙,科学上网,梯子。非永久免费梯子,官网,非永久免费VPN,非免费机场!VPN China。按量计费机场,按量付费机场,不限时机场。机场节点订阅,节点分享,Clash节点,V2ray节点,非免费节点。适用ClashForWindows,V2rayN,singbox,nekobox,小火箭等代理软件🚀🚁
Rust GUI components for building fantastic cross-platform desktop application by using GPUI.
NSH runs modules written in TSX — natively on iOS, Android, macOS, Windows, Linux, GPUI and the web
Fine-tune LLMs from one YAML. Layer streaming trains an 8B model on a 4 GB laptop GPU.
Production-grade platform for building agentic IM bots - 生产级多平台智能机器人开发平台/ Agent、知识库编排、插件系统 / Bots for Discord / Slack / LINE / Telegram / WeChat(企业微信, 企微智能机器人, 公众号) / 飞书 / 钉钉 / QQ / Matrix e.g. Integrated with ChatGPT(GPT), DeepSeek, Dify, n8n, Langflow, Coze, Claude, Gemini, GLM, Ollama, SiliconFlow, Kimi, openclaw / hermes agent, deerflow
ScopeSentry-Cyberspace mapping, subdomain enumeration, port scanning, sensitive information discovery, vulnerability scanning, distributed nodes
The recursive internet scanner for hackers. 🧡
Make videos on the computer you already own. FreeVideo runs MiniMax H3 in as little as 8 GB of VRAM and 16 GB of RAM, and adapts its acceleration path to your hardware.
Unlock Google 11's Gemini "Rambler" Natural Voice Dictation & AI Writing Tools on ANY Android phone. No root required.
Garry's Opinionated OpenClaw/Hermes Agent Brain
Geometry-Aware Monocular RGB Planning
Deep Learning Book Chinese Translation
🔥 官方推荐 🔥 RuoYi-Vue 全新 Pro 版本,优化重构所有功能。基于 Spring Boot + MyBatis Plus + Vue & Element 实现的后台管理系统 + 微信小程序,支持 RBAC 动态权限、数据权限、SaaS 多租户、Flowable 工作流、三方登录、支付、短信、商城、CRM、ERP、MES、IM、AI 大模型、IoT 物联网等功能。你的 ⭐️ Star ⭐️,是作者生发的动力!
Implement one small API on SQLite in any language; we benchmark it on a $12 server (1 vCPU, 2 GB)
Xolotl (formerly Strata-DSV4.1): DeepSeek-V4.1-Flash, all 552B, on one gaming PC (RTX 4090 + 90 GB RAM) at ~27 tok/s, plus GLM-5.3-Flash, Qwen3.8-Flash-Next and Qwen3.8-27B. A heavily extended fork of Strata: MoE experts tiered across VRAM, pinned RAM and NVMe; OpenAI + Anthropic API.
A Game Boy, Game Boy Color and GBA player for Android handhelds with two screens or one, with a live companion for Pokémon games and ROM hacks.
HexaShell: a thin harness in which a coding agent drives a robot from RGB images and writes its own tools; frozen tools for the 42 RoboDojo tasks
Turkish voice design with VoxCPM2 on a 4 GB GPU: describe a voice, pick a variant, read Turkish text. Patched audio.cpp, local UI, 7 synthetic voices.
SMG GBA Project
A Claude Code skill bundle for bug hunting and external red-team work - 82 skills, 15 slash commands, 681 disclosed-report patterns curated across 24 core vulnerability classes, plus enterprise identity + infrastructure attack matrices.
Community curated list of templates for the nuclei engine to find security vulnerabilities.
Multi-platform auto-proxy client, supporting Sing-box, X-ray, TUIC, Hysteria, Reality, Trojan, SSH etc. It’s an open-source, secure and ad-free.
Serve large Qwen models fast on the GPUs you actually own. Qwen3.8-27B on a single 24 GB card with vLLM: 127 tok/s single-user (381 when the answer quotes the prompt), ~1,035 tok/s at 64 concurrent, 150k-262k context. vLLM patches, requant pipeline, benchmarks.
2026年稳定高速VPN推荐:40款性价比翻墙梯子完全指南,每周核查跑路与价格。翻墙梯子/科学上网/机场测评,解锁 YouTube、Netflix、TikTok、ChatGPT,适用 Clash、V2ray、小火箭、sing-box,支持 Windows、Mac、iOS、安卓、路由器
Fish Audio Drama 3 text-to-speech on the TensorFold native CUDA engine for DGX Spark (GB10): recipe, engine patches, benchmarks. Built with Fish Audio.
Open PS5 firmware 14.10 research: Lua userland experiments, syscall and attack-surface analysis. Kernel read/write and a complete exploit are not demonstrated.
Open source software that helps you create and deploy high-frequency crypto trading bots
Gemma 4 26B-A4B inference in ~2 GB of RAM on any M-series MacBook
Self-hosted homelab infrastructure visualizer — interactive network diagram with live status monitoring
A 2.78-trillion-parameter Kimi K3 running inference on a single CPU in 8.24 GB of RAM. Portable C99: no BLAS, no framework, no GPU.
Free proxy node aggregator. Daily updated vmess/vless/trojan/ss. 免费 代理节点 翻墙 梯子 科学上网 机场订阅 网络加速 VPN 免翻墙每日更新 一键导入
A curated list of various bug bounty tools
Ninja Gaiden Black (Xbox, 2005) statically recompiled into native Linux and Windows builds, drawn by its own Vulkan renderer. A steady 60 fps, widescreen and ultrawide, keyboard and mouse, Steam Deck ready. Bring your own copy of the game.
YouTube view bot: multi-threaded view inflation, proxy rotation, watch-time retention, and randomized user-agent/device fingerprinting per view.
A 100 GB live-synced database in one Durable Object: Rust page engine in R2, incremental live queries, Linear-style demo
reconFTW is a tool designed to perform automated recon on a target domain by running the best set of tools to perform scanning and finding out vulnerabilities
THE MOST POWERFUL OFFENSIVE AI AGENT FOR PENTESTING , OSINT , AND ETHICAL HACKING
Shows what's playing in Home Assistant as a calm picture on Android TV, with Sendspin support.
A curated list of awesome search engines useful during Penetration testing, Vulnerability assessments, Red/Blue Team operations, Bug Bounty and more
Persistent memory for Claude Code, Codex and 6 other AI agents, stored as plain markdown in your Obsidian vault. New: Crew, one shared logbook so Claude Code, Codex, Cursor and Grok Bot know what each other did. 48 commands: semantic search, self-rewriting notes, key-less research, scheduled agents.
Label Studio is a multi-type data labeling and annotation tool with standardized output format
A bespoke, Game Boy-centric CFW for the RG SP.
DeepSeek-V4.1-Flash-1017 EXL3 2.9bpw on 2x NVIDIA DGX Spark (GB10) with TensorFold: 2.1-2.3x the vLLM kit's decode speed.
番茄小说下载器不精简版
Large list of handpicked color names 🌈
The Web Security Testing Guide is a comprehensive Open Source guide to testing the security of web applications and web services.
🏖️ A serverless, open-source file hosting solution built on Cloudflare. Supports image hosting, secure file storage, and personal cloud drive capabilities.
A markdown editor — and the bridge to your LLM. Local-first, MIT, ~15 MB. Bundled MCP server lets Claude Code / Codex / Cursor drive your vault directly. 14 AI providers BYOK.
httpx is a fast and multi-purpose HTTP toolkit that allows running multiple probes using the retryablehttp library.
Good internal base With Model Changer Always update
DeepSeek-V4.1-Flash-1017 (552B MoE) on 2x RTX 3090 + 128 GB RAM: Rust + CUDA engine and 2-bit quant. 30-37 tok/s decode.
[ECCV 2026 Best Paper Award Candidate] LingBot-Map: Geometric Context Transformer for Streaming 3D Reconstruction
gbOS86: Googlebook OS on ordinary Intel laptops (6th gen+): CPU-instruction emulation, patched libs, boot-time hardware detection, live/installer USB.