Research and development (R&D) is crucial for the enhancement of industrial productivity, especially in the AI era, where the core aspects of R&D are mainly focused on data and models. We are committed to automating these high-value generic R&D processes through R&D-Agent, which lets AI drive data-driven AI. 🔗https://aka.ms/RD-Agent-Tech-Report
Official Inference Code of EAGLE-4
A fast, keyboard-driven TUI API client & webhook inspector. No cloud, no Electron, 100% local-first.
Open-source alternative to Opus Clip, Vidyo.ai, Klap & SubMagic. Turn long-form YouTube videos into viral 9:16 shorts using LLM highlight detection, Whisper transcription, and auto vertical cropping — free, no watermarks, no per-clip credits.
[NeurIPS 2026] DataFlex: A Unified Benchmark and Evaluation Platform for Data-Centric Training of Large Language Models
Native vLLM serving for Jev decision models
Let your AI agents call your phone over VoIP, and call them back. Your phone rings like a real call, even locked; a voice AI talks it through and returns the transcript. Self-hosted, open source, no phone numbers or Twilio.
Connect any AI copilot or autonomous agent to your enterprise tools — through a single, secure MCP/Agent gateway with built-in identity, access control, and full observability.
Find the ghost tokens. Fix them. Survive compaction. Avoid context quality decay.
CRAB (Clean Rebuild of Anything's Behavior): rebuild software from behavior, not source code. An MCP-powered toolkit for black-box observation, clean-room reimplementation, and behavioral verification.
AI-powered travel hacking and search with cash, points, miles, and award flights. Drop-in skills and MCP servers for Claude, Codex, and OpenCode.
JevStyle: local decision models for AI agents. Classify, route and score with a probability per option. systemone-compatible local server, MCP tools, Claude Code guard. Open weights, Apache-2.0.
MCP server integration for DaVinci Resolve Studio
PhysAg3D: Bridging 3D Generative and Agentic Models for High-Fidelity and Physically-Interactive Assets
[SIGMOD'27] Easy Data Preparation with latest LLMs-based Operators and Pipelines.
JiuwenSwarm is an intelligent AI Agent built on openJiuwen. It extends the powerful capabilities of large language models directly to your fingertips through various communication apps you use daily.
A Claude Code skill bundle for bug hunting and external red-team work - 82 skills, 15 slash commands, 681 disclosed-report patterns curated across 24 core vulnerability classes, plus enterprise identity + infrastructure attack matrices.
Open-source State-of-the-art TTS model which runs on a CPU 😻
A Text-to-SQL agent that diagnoses failed queries and promotes validated, versioned query-writing skill updates instead of changing model weights.
磁盘空间变化对比轻量小工具;A lightweight disk space change comparison tool.
A production-grade, high-efficiency Agentic RL framework for frontier LLMs with validated recipes.
[WACV 2027] ComplexSync: High-Fidelity and Real-Time Lip Sync in Complex Scenarios
PanWatch — AI stock monitoring for A-shares, HK & US markets, powered by TradingAgents. Portfolio insights, real-time alerts & automated reports.|盯盘侠:覆盖 A股/港股/美股的 AI 盯盘、持仓分析、实时提醒与自动报告。
Qwen3.5-2B with a filesystem compiled and inserted directly into its weights.
Musicdl: A lightweight music downloader written in pure python. (轻量级无损音乐下载器,支持数十个音乐/有声读物平台,例如网易云音乐,QQ音乐,酷狗音乐,酷我音乐,咪咕音乐,千千静听,汽水音乐,Bilibili,街声,喜马拉雅,懒人听书,荔枝FM,蜻蜓FM,JOOX,TIDAL,YouTube,Apple Music,Spotify,Qobuz,SoundCloud等主流音乐平台)
:movie_camera: Python and OpenCV-based scene cut/transition detection program & library.
The Web framework for perfectionists with deadlines.
Convert PDF to markdown + JSON quickly with high accuracy
A framework for continual learning and recursive self-improvement through harness adapter evolution.
Naive-N0.5-Flash is an open-weight 309B MoE model with 15.5B active parameters, built for coding and AI R&D.
A lightweight system to automate posting to all of your socials from one dashboard
Google Search Console Insights with Claude AI for SEOs
Framework for training small decision models through play, using Red Alert 2 and Tank War as test environments.
联想 ThinkSmart View(CD-18781Y)语音助手系统镜像:基于 rickx/ubuntu-thinksmart,开机直接进语音界面;含中英文刷机教程(写给人和 AI 助手)、出镜像脚本和设备修复服务|Voice-assistant system image for Lenovo ThinkSmart View (CD-18781Y), based on rickx/ubuntu-thinksmart: boots straight into the voice UI; flashing guide in Chinese and English (for people and AI agents), image build scripts and device fixes
Official repository for LTX-Video
MiniMax Hailuo 02 API (`MiniMax-Hailuo-02`): model id, pricing, parameters and copy-paste examples for the OpenAI-compatible APIMart endpoint.
Official codebase for "Multi-Token Prediction Done Right"
Handheld Raspberry Pi flight tracker for a young plane-spotter: live map, plane photos and routes, a logbook, ADS-B antenna fallback and a 3D-printed case
ChatDev 2.0: Dev All through LLM-powered Multi-Agent Collaboration
GenBI (Generative BI) for AI agents, an open-source, governed text-to-SQL through an open context layer that turns natural-language questions into trusted dashboards, charts, and SQL across 20+ data sources, such as BigQuery, Snowflake, PostgreSQL, ClickHouse, Amazon Redshift, Databricks and more.
Agent skill that directs scripts for TTS (ElevenLabs v3/v4 tags): rewrite for the ear, tag budget, free linter, A/B tester. | اسکیلی برای کارگردانی متن قبل از TTS: بازنویسی برای گوش، تگگذاری بدون زیادهروی، لینتر رایگان و تست A/B. فارسیمحور.
AI job application bot: automated job search and job hunting, auto-apply to resume-matching jobs while you sleep.
Official Hex-Rays IDA MCP Server.
Terrain Grounded Imagination: a recipe for training humanoid locomotion on sparse and rough terrain through a world model with differentiable heightmap lookups.
WeVisCut is a 9B tool-using agent that turns knowledge articles and narration scripts into multi-minute videos through planning, asset retrieval, editing, inspection, and repair. Built on Qwen3.5, it operates in a stateful 12-tool workflow and includes an editable Studio runtime.
Minimal CLI coding agent by Mistral
MiniMax Hailuo 02 prompts: tested image prompt recipes with example renders, aspect ratios, the model id (`MiniMax-Hailuo-02`) and the cost per image.
AI-native video production toolkit for Claude Code
Getting ADB, root, backups and custom apps onto locked-down Android photo frames (Skylight Calendar, BIUFRAME)
Build and share delightful machine learning apps, all in Python. 🌟 Star to support our work!
VLX-Seek is an efficient VLM that grounds fine-grained visual targets with high precision in wild environments.
—
Fewer Tokens, Better Action: GPT-6 Astra Robot Agents with 14% Higher Success but 65% Fewer Tokens
Same take, new cast: open-weights video character recasting on local ComfyUI with SAM 3.1, FLUX 3 Image Klein and SCAIL-2. Built for #OpenSourceHiggsfield.
Self-hosted tabular ML that doesn't lie. Leakage Guard, conformal intervals, profit-optimized thresholds, counterfactuals & instant REST/MCP API.
2026智慧树刷课脚本 基于Python Playwright的自动化程序 [有免安装版]
Learn a codebase by re-deriving its design decisions yourself. A Claude Code skill.
Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels
A simple and lightweight app for running Windows games using UMU-Launcher
Seedream 5.0 Flash prompts: eight tested image prompt recipes with example renders, aspect ratios, the model id (`seedream-5-0-flash`) and the cost per image.