#
Starred repositories
RooVetGit / Roo-Code
Forked from cline/clineRoo Code (prev. Roo Cline) gives you a whole dev team of AI agents in your code editor.
Lightweight coding agent that runs in your terminal
Context7 MCP Server -- Up-to-date code documentation for LLMs and AI code editors
FB (Facebook) + GEMM (General Matrix-Matrix Multiplication) - https://code.fb.com/ml-applications/fbgemm/
nvidia-modelopt is a unified library of state-of-the-art model optimization techniques like quantization, pruning, distillation, speculative decoding, etc. It compresses deep learning models for do…
Optimized primitives for collective multi-GPU communication
A Unity MCP server that allows MCP clients like Claude Desktop or Cursor to perform Unity Editor actions.
Model Context Protocol Servers
Monitor browser logs directly from Cursor and other MCP compatible IDEs.
An open-source dashboard for Cursor.sh IDE. Log AI code generations, track usage, and control AI models (including local ones). Run locally or use upcoming hosted version.
llama.cpp fork with additional SOTA quants and improved performance
🚀 The fast, Pythonic way to build MCP servers and clients
FULL v0, Cursor, Manus, Same.dev, Lovable, Devin, Replit Agent, Windsurf Agent & VSCode Agent (And other Open Sourced) System Prompts, Tools & AI Models.
Official repository for the paper "LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code"
A framework for few-shot evaluation of language models.
A streamlined and customizable framework for efficient large model evaluation and performance benchmarking
FlashMLA: Efficient MLA decoding kernels
Tools for merging pretrained large language models.
A lightweight, powerful framework for multi-agent workflows
Code for the paper "Evaluating Large Language Models Trained on Code"
LM inference server implementation based on *.cpp.
A (nicer) tokenizer you want to use for model inference and training: with all known peventable gotchas normalized or auto-fixed.
Production ready LLM model compression/quantization toolkit with hw accelerated inference support for both cpu/gpu via HF, vLLM, and SGLang.
Transformers-compatible library for applying various compression algorithms to LLMs for optimized deployment with vLLM
[CVPR 2024 Oral] InternVL Family: A Pioneering Open-Source Alternative to GPT-4o. 接近GPT-4o表现的开源多模态对话模型