Stars
This repository contains the official implementation of "FastVLM: Efficient Vision Encoding for Vision Language Models" - CVPR 2025
Recursive-Open-Meta-Agent v0.1 (Beta). A meta-agent framework to build high-performance multi-agent systems.
Portable file server with accelerated resumable uploads, dedup, WebDAV, FTP, TFTP, zeroconf, media indexer, thumbnails++ all in one file, no deps
Letta is the platform for building stateful agents: open AI with advanced memory that can learn and self-improve over time.
🕵️♂️ TUI for sniffing network traffic using eBPF on Linux
Grist is the evolution of spreadsheets.
CoexistAI is a modular, developer-friendly research assistant framework . It enables you to build, search, summarize, and automate research workflows using LLMs, web search, Reddit, YouTube, and ma…
Code of fine-tuning neural sparse models and training from scratch. #SIGIR2025
HelixDB is an open-source graph-vector database built from scratch in Rust.
AI orchestration framework to build customizable, production-ready LLM applications. Connect components (models, vector DBs, file converters) to pipelines or agents that can interact with your data…
✨ AI interface for tinkerers (Ollama, Haystack RAG, Python)
Write like nobody’s watching. Because they aren't.
This project automates the process of collecting, rating, and sending newsletter articles using n8n. It integrates RSS feeds, AI-based article rating and summarization, and automated email delivery.
Identity-Aware Tunneled Reverse Proxy Server with Dashboard UI
all of the workflows of n8n i could find (also from the site itself)
Abbey is a self-hosted configurable AI interface with workspaces, document chats, YouTube chats, and more. Find our hosted version at https://abbey.us.ai.
Generate a wiki for your research topic, sourcing from the web and your docs.
AI chatbot for Matrix with infinite personalties, using ollama
The Official Model Context Protocol (MCP) server for Kagi search & other tools.
Open Source Deep Research Alternative to Reason and Search on Private Data. Written in Python.
Milvus is a high-performance, cloud-native vector database built for scalable vector ANN search
All-in-one LLM CLI tool featuring Shell Assistant, Chat-REPL, RAG, AI Tools & Agents, with access to OpenAI, Claude, Gemini, Ollama, Groq, and more.