Search

🐋 Run 284B DeepSeek-V4-Flash (<59GiB RAM, 44.4 tok/s pp, 9.1-14.7 tok/s) & 30B MoEs (25-111 tok/s) on a pure 64GB CPU at 256K ctx (0 GPUs). SwiGLU-H2, Stockfish-PVS L3 & 13ms Claude Code restore.

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.

Apple's open-source XNU kernel on Android phones: its own loader and drivers, Metal on the Adreno. Runs from RAM; a reboot gives the phone back to Android.

Xolotl: run DeepSeek-V4.1-Flash, all 552B, as a local coding agent on one gaming PC (RTX 4090 + 110 GB RAM): ~33 tok/s on agent tasks, ~41 in fast mode, 512K context. Plus GLM-5.3-Flash, Qwen3.8 and LFM3-Omni. MoE experts tiered across VRAM, pinned RAM and NVMe, with Xolotl Direct NVMe->VRAM. OpenAI + Anthropic API.

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

A free, open-source Stardock Fences alternative for Windows 11 and 10. Glass fences, live folders, tabs, Peek. Rust, a few MB of RAM at rest, 11 languages, portable.

A native Discord client of about 100~ MB RAM, whose entire interface is a theme file.

Gemma 4 26B-A4B inference in ~2 GB of RAM on any M-series MacBook

🦔 Fast, lightweight & schema-less search backend. An alternative to Elasticsearch that runs on a few MBs of RAM.

:ram: Practical functional Javascript

Get your data in RAM. Get compute close to data. Enjoy the performance.

Recursive version of fs.readdir with small RAM & CPU footprint.

An ultra-lightweight(ROM<1.6K, RAM<0.3k), high-performance C/C++ log library. | 一款超轻量级(ROM<1.6K, RAM<0.3k)、高性能的 C/C++ 日志库

LuatOS -- Powerful embedded Lua Engine for IoT devices, with many components and low memory requirements (16K RAM, 128K Flash)

Improved YouTube chat with CPU/RAM optimizations, customization options, and cutting-edge features!

xfrpc 是一个轻量级的 FRP 客户端，完美兼容 frps，采用 C 语言实现，专为 OpenWRT 和物联网等资源受限系统优化设计。它针对 ROM 和 RAM 空间有限的设备，提供高效的内网穿透解决方案。xfrpc 集成了xDPI（深度包检测）功能，增强了安全性，有效防止内网穿透中因恶意嗅探导致的安全威胁，确保数据传输和网络访问的可靠保护。技术交流QQ群 331230369

Limits Discord, Chrome, OBS or Any Applications RAM Usage In 1 Click.

A framework to study AI models in Reasoning, Alignment, and use of Memory (RAM).

Low-level networking. HTTP/TCP/IP/Ethernet/RMII/MDIO/UDP/NTP/DHCP/DNS blabla. Runs on a couple kB of RAM

Emulators running NES, SNES, GameBoy, GameGear, Master System, Genesis / Megadrive, Neo Geo Pocket, WonderSwan, Lynx, Atari7800, Atari2600, PC-Engine, Coleco, SG1000, GX4000 and MSX on ESP32 with 320KB of RAM

一款轻量级、高度可定制的 Windows桌面和任务栏硬件性能监控工具，支持监测 CPU、GPU、内存、磁盘、网速、FPS 计数、插件扩展及内存清理。A lightweight, customizable hardware monitor for the Windows desktop & taskbar. Features CPU/GPU/RAM/Network monitoring, FPS counter, plugin support, and memory optimization.

A deterministic genomics engine with a compact memory footprint. Run whole-genome workloads in as little as 100 MB RAM. Built in Rust.

ESP32-C3 + OLED desk monitor: live CPU/GPU/RAM/disk stats from a Windows/Linux companion app, and 12 animated retro clock styles when the PC sleeps.

Noodle provides primitive modular functions for convolution layer, dense layer, pooling and activations. It allows streaming the intermediate activations, weights, and biases from/to SD/FFat/SD_MMC filesystems to overcome RAM limitations.

Lightweight multi-format artifact registry. 13 formats: Docker, Maven, npm, PyPI, Cargo, Go, NuGet, RubyGems, Terraform, Conan, Ansible Galaxy, Dart Pub, and Raw. Single binary, zero dependencies, <100MB RAM.

Fast, lightweight Firecrawl/Tavily alternative in Rust. Web scraper, crawler & search API with MCP server for AI agents. Drop-in Firecrawl-compatible API (/scrape, /crawl, /search). 2.3x faster than Tavily, 1.5x faster than Firecrawl in 1K-URL benchmarks. 6 MB RAM, single binary. Self-host or use managed cloud.

Lightweight Rust Agent only use 50MB RAM, but Claude Code Plugin compatible, Dynamic Workflow, Goal, Artifacts, Free Web Search, full feature and better support!

Library for reducing tail latency in RAM reads

The cost efficiency of S3 with the speed of local RAM. A multi-tenant vector and full-text search engine featuring a tiered RAM -> NVMe -> S3 architecture for microsecond latency on top of object storage

Free Wallpaper Engine Steam: download PC windows 11, steam workshop downloader bypass, high CPU RAM usage fix, android apk mobile sync. 4k anime live wallpaper, black screen crash on startup, corsair icue razer chroma integration, ui32.exe error solution, scene web wallpaper, mobile connection pin, custom playlist screensaver,