fastllm

fastllm是后端无依赖的高性能大模型推理库。同时支持张量并行推理稠密模型和混合模式推理MOE模型，任意10G以上显卡即可推理满血DeepSeek。双路9004/9005服务器+单显卡部署DeepSeek满血满精度原版模型，单并发20tps；INT4量化模型单并发30tps，多并发可达60+。

Starward

Game Launcher for miHoYo - 米家游戏启动器

ultimate-react-course

Starter files, final projects, and FAQ for my Ultimate React course

JimsGarage

Homelab Goodies

van

🍦 VanJS: World's smallest reactive UI framework. Incredibly Powerful, Insanely Small - Everyone can build a useful UI app in an hour.

llm-numbers

Numbers every LLM developer should know

GodMode

AI Chat Browser: Fast, Full webapp access to ChatGPT / Claude / Bard / Bing / Llama2! I use this 20 times a day.

CapsWriter-Offline

CapsWriter 的离线版，一个好用的 PC 端的语音输入工具

WuKongIM

More than just IM 不只是即时通讯( IM )

RT-DETR

[CVPR 2024] Official RT-DETR (RTDETR paddle pytorch), Real-Time DEtection TRansformer, DETRs Beat YOLOs on Real-time Object Detection. 🔥 🔥 🔥

geekai

AI 助手全套开源解决方案，自带运营管理后台，开箱即用。集成了 ChatGPT, Azure, ChatGLM,讯飞星火，文心一言等多个平台的大语言模型。支持 MJ AI 绘画，Stable Diffusion AI 绘画，微博热搜等插件工具。采用 Go + Vue3 + element-plus 实现。

tree-of-thoughts

Plug in and Play Implementation of Tree of Thoughts: Deliberate Problem Solving with Large Language Models that Elevates Model Reasoning by atleast 70%