Repository History
186 repositories tagged with AI
Headroom: Drastically Reduce LLM Token Usage for AI Agents
Headroom is an innovative context compression layer for AI agents, designed to significantly reduce token usage for LLMs. It achieves 60-95% fewer tokens across various inputs like tool outputs, logs, files, and RAG chunks, all while preserving answer accuracy. This powerful tool enhances efficiency and cost-effectiveness for AI interactions.
Voicebox: The Open-Source AI Voice Studio for Cloning and Dictation
Voicebox is an innovative open-source AI voice studio that allows users to clone voices, generate speech in multiple languages, and dictate into any application. It provides a comprehensive, local-first voice I/O stack, offering a powerful alternative to cloud-based solutions. This tool ensures complete privacy and control over your voice data, running entirely on your local machine.
Dexter: An Autonomous Agent for Deep Financial Research
Dexter is an autonomous financial research agent designed to think, plan, and learn while performing analysis. It leverages task planning, self-reflection, and real-time market data to tackle complex financial questions. This project provides a powerful tool for in-depth financial exploration, emphasizing its educational and informational purposes.
PixelRAG: Pixel-Native Search for Visual Retrieval-Augmented Generation
PixelRAG revolutionizes search by enabling pixel-native retrieval, moving beyond traditional text parsing. It renders documents as screenshots, preserving visual context like tables and charts, which is crucial for accurate answers from reader models. This allows for searching any document based on its visual appearance, not just its textual content.
GLM-5: Flagship Models for Long-Horizon Agentic Engineering
GLM-5 is a series of flagship models, including GLM-5.2, GLM-5.1, and GLM-5, developed by zai-org for complex systems engineering and long-horizon agentic tasks. These models offer advanced coding capabilities, impressive context lengths, and state-of-the-art performance on various benchmarks. They are designed to sustain effective problem-solving over extended sessions through iterative reasoning and strategy revision.
Synara: A Local-First Desktop App for AI-Powered Coding
Synara is a local-first desktop application designed to streamline coding with various AI agents and subscriptions. It integrates chats, terminals, browser previews, and Git operations into a single focused workspace, eliminating the need to juggle multiple windows. This app empowers developers to leverage their existing AI tools for parallel work across projects while maintaining a private, local environment.
Qwen3-VL: A Powerful Multimodal Large Language Model Series
Qwen3-VL is a cutting-edge multimodal large language model series from Alibaba Cloud's Qwen team. It offers significant advancements in visual and text understanding, extended context length, and enhanced agent capabilities. This model is designed for flexible deployment, scaling from edge to cloud.
Understand-Anything: Interactive Knowledge Graphs for Codebases
Understand-Anything is an innovative open-source project that transforms any codebase or knowledge base into an interactive knowledge graph. It allows developers to visually explore, search, and query their projects, making complex systems easier to comprehend. This tool integrates seamlessly with popular AI coding platforms like Claude Code, Codex, and Copilot.
AutoHedge: Build Your Autonomous AI Hedge Fund with Swarm Intelligence
AutoHedge is an enterprise-grade autonomous agent hedge fund that leverages swarm intelligence and specialized AI agents. This powerful Python project automates end-to-end market analysis, risk management, and trade execution. It allows users to build and deploy their own AI-driven trading strategies with minimal human intervention.
Open-Higgsfield-AI: Free, Self-Hosted AI Image Generation & Cinema Studio
Open-Higgsfield-AI offers an open-source, self-hosted alternative for AI image generation and a cinema studio. It provides access to over 20 models, including Flux, SDXL, Midjourney, and Ideogram, allowing users to create stunning visuals and cinematic content. This MIT-licensed project is fully customizable and designed for local operation.
Cabinet: An AI-First Knowledge Base and Startup OS
Cabinet is an AI-first knowledge base and startup OS designed to keep your data local and private. It allows you to build a custom AI team with agents that remember context, manage tasks, and automate workflows. This self-hosted solution ensures your information remains under your control, without vendor lock-in.
Compound Engineering Plugin: AI Skills for Smarter, More Efficient Development
The Compound Engineering plugin introduces a revolutionary approach to software development, leveraging AI agents to make each unit of engineering work easier than the last. It focuses on thorough planning, insightful reviews, and codifying knowledge to prevent technical debt. This plugin integrates with popular AI coding environments like Claude Code, Cursor, and Copilot, transforming traditional development into a compounding process of continuous improvement.