AI Agents & Automation

AgentBench is a comprehensive benchmark designed to evaluate Large Language Models (LLMs) as agents across a diverse spectrum of environments. It encompasses 8 distinct environments, including 5 newly created domains like Operating System (OS), Database (DB), Knowledge Graph (KG), Digital Card Game (DCG), and Lateral Thinking Puzzles (LTP), alongside 3 recompiled from published datasets (House-Holding, Web Shopping, Web Browsing). The platform offers both Dev and Test splits for each dataset, requiring LLMs to generate responses thousands of times for thorough evaluation. AgentBench also introduces VisualAgentBench for evaluating and training visual foundation agents based on large multimodal models (LMMs), covering embodied, GUI, and visual design environments. It supports quick setup using Docker Compose and provides benchmarking results via a leaderboard.

Vanim is an AI-powered English speaking tutor designed to help users master English with confidence. It offers a 100% free, offline experience with no signup or personal data collection, ensuring privacy. The tool focuses on spoken practice, moving beyond typing and multiple-choice questions, with features like structured learning paths from beginner to advanced, real conversations with AI on various topics, and instant feedback on grammar, vocabulary, pronunciation, and fluency. Users can practice real-world English scenarios, including interviews, office small talk, and casual conversations, making it ideal for job seekers, students, professionals, and travelers.

Buyutech is a full-stack perception company specializing in camera-based sensing technologies for automotive, defense, and industrial mobility. They develop complete technology stacks, from photon to real-time perception, enabling safe and intelligent movement for vehicles, robots, and autonomous systems in various environments. Their offerings include core automotive products like analog and digital rear-view cameras, digital side mirror systems, occupant and driver monitoring systems, and surround-view camera systems. For defense and aerospace, they provide mission-critical terminal solutions, perception for aerial platforms, and situational awareness systems. Industrial mobility solutions include stereo depth cameras, 360° perception systems, AI-driven navigation modules, and blind-spot detection cameras. Buyutech integrates hardware, imaging pipelines, edge AI, fusion, and high-volume camera production to deliver highly reliable perception.

EASYChatGPT is an open-source desktop application project designed to facilitate developer access to ChatGPT. It provides a straightforward way for users to interact with ChatGPT's interface directly from their desktop environment, requiring only a personal API key. The project emphasizes ease of use, with a two-step setup process for installation and conversation initiation. It's particularly useful for developers who want to experiment with ChatGPT functionalities without needing to rely on web interfaces or complex setups. The tool currently supports single-turn conversations and requires users to replace the API key in the configuration file. It's important to note that this is a personal project and not an official OpenAI product.

flockx offers specialized AI agents designed to act as marketing, sales, and operations specialists, enabling businesses to scale efficiently without the need for additional human hires. These AI teams are ready in just one minute, providing a quick solution for creators and creative professionals. The platform emphasizes business automation and workflow intelligence through multi-agent systems, offering custom AI agent building capabilities. It aims to be an alternative to tools like Zapier and Make.com for automating small business operations, including customer service. flockx is part of the Fetch.ai ecosystem and is trusted by over 3,000 businesses worldwide.

Feathr is a scalable, unified data and AI engineering platform widely used in production at LinkedIn and now an open-source project under the LF AI & Data Foundation. It allows users to define data and feature transformations using Pythonic APIs, register these transformations, and share them across teams. Particularly useful for AI modeling, Feathr automatically computes and joins feature transformations to training data with point-in-time correctness to prevent data leakage. It supports materializing and deploying features for online production use, offers native cloud integration with scalable architecture, and has been battle-tested for over six years. Feathr handles billions of rows and petabyte-scale data with built-in optimizations, providing rich transformation APIs including time-based aggregations and sliding window joins. It also features a built-in registry for feature reuse and an intuitive UI for searching and exploring features and their lineages.

k8m is a lightweight, cross-platform Mini Kubernetes AI Dashboard designed to streamline cluster management. Built on AMIS and using kom as a Kubernetes API client, it integrates AI capabilities like Qwen2.5-Coder-7B and DeepSeek-R1-Distill-Qwen-7B for intelligent analysis, YAML translation, and log AI diagnostics. It supports multi-cluster management with heart-beat detection, automated reconnection, and granular permission control for users and groups. Key features include a plugin-based architecture, MCP integration for large model tool calls, and advanced security with MCP permission integration. It also offers Pod file and running management, API access, cluster inspection, k8s Event forwarding, CRD management, and a Helm market. The tool is fully open-source, supports multiple architectures and databases, and can be deployed as a single executable, making it highly efficient and easy to use for Kubernetes operations.

LLM-Pruner is a cutting-edge tool designed for the structural pruning of large language models (LLMs), as presented at NeurIPS 2023. It enables users to compress LLMs to any desired size while retaining their original multi-task solving abilities. The tool emphasizes task-agnostic compression, requiring minimal training corpus (e.g., 50k Alpaca samples for post-training) and offering efficient compression times, with pruning taking approximately 3 minutes and post-training around 3 hours. LLM-Pruner supports a wide range of popular LLMs, including Llama-3/3.1, Llama-2, LLaMA, BLOOM, Vicuna, Baichuan, and TinyLlama. It features an automatic structural pruning process, aiming for minimal human effort, and provides detailed instructions for discovery, estimation, and recovery stages of pruning, along with evaluation using lm-evaluation-harness.

Certainly offers AI agents designed to automate customer conversations across multiple channels, including web, WhatsApp, email, voice, and SMS. This platform is built for enterprise customer experience (CX) teams looking to streamline operations and provide instant support. Key features include no-code deployment and multi-lingual capabilities, making it accessible and versatile for global businesses. Certainly's AI agents can handle customer service, sales, and internal support, aiming to enhance efficiency and customer satisfaction by automating routine interactions and providing quick, accurate responses.

LingChat is an AI chat companion that integrates emotional expressions into its GPT conversations. It utilizes a self-trained AI emotion recognition model to determine the AI's emotional state during each dialogue, influencing its expressions, actions, and chat bubble styles. The tool offers permanent memory for each saved conversation, allowing for consistent and personalized dialogue styles. Users can customize characters, import scripts for multi-role conversations, and even enable visual perception for the AI to interpret screen activity. LingChat supports Windows, Linux, and macOS, including 32-bit Windows systems and older CPUs, making it accessible to a wide range of users.

OMG is an advanced open-source framework designed for occlusion-friendly personalized multi-concept generation within diffusion models, as presented at ECCV 2024. It allows users to generate complex images featuring multiple characters and styles, integrating seamlessly with LoRAs from Civitai.com and InstantID for single-image ID personalization. The tool also supports ControlNet for layout control and various style LoRAs. OMG is built on Python 3.10.6 with PyTorch 2.0.1 and torchvision 0.15.2, requiring specific model downloads for its functionality, including Stable Diffusion XL and various ControlNet and LoRA checkpoints. It offers flexible usage through command-line inference scripts for both LoRA and InstantID workflows.

notion_widgets is an open-source collection of HTML widgets designed to enhance Notion.so pages. Users can embed these widgets to add various interactive and functional elements to their Notion workspaces, customizing their experience beyond Notion's native capabilities. The project includes a diverse range of widgets such as calendars, countdown timers, currency converters, weather displays, and more, providing practical tools for organization and productivity. By offering these HTML-based solutions, notion_widgets empowers users to create more dynamic and personalized Notion environments, making it a valuable resource for those looking to extend their Notion functionality.

Nanoswarm offers personal AI agents specifically designed for Telegram, allowing users to easily set up and customize their bots. The platform emphasizes personalization, enabling users to tweak the AI's personality, role, and tone to make each OpenClaw bot uniquely their own. It supports various frontier AI models, giving users flexibility in their agent's capabilities. Security and privacy are core features, with tokens encrypted at rest to ensure user data remains private. Furthermore, Nanoswarm guarantees continuous operation with dedicated infrastructure that keeps bots online 24/7, providing reliable and always-available assistance within Telegram.

VirtualWife is a virtual digital human project designed to provide companionship and emotional support. Currently in its incubation phase, the project aims to create a virtual digital human with its own "soul" that users can interact with like a friend. Key features include one-click Docker deployment, support for Linux/Windows/MacOS, customizable character settings, and the ability to change character models from VRM markets. It offers long and short-term memory functions, multi-LLM model switching (including private models like Ollama), and supports text-driven expressions and actions. The tool also integrates with Bilibili for live streaming and enables voice conversations in Chinese, with support for Edge (Microsoft) and Bert-VITS2 voice switching for faster response times through streaming data.

antigravity-agent is an open-source tool designed for the effortless management of multiple Antigravity accounts. It allows users to quickly switch between accounts with a single click, eliminating the need for repetitive logins. The software automatically identifies and saves current account data, and offers secure backup functionality through password-encrypted export of account configurations, facilitating cross-device migration. A VSCode extension is also available, enabling users to switch accounts and view model quotas directly within their editor. The tool emphasizes security, distributing only through official GitHub Releases to protect sensitive account information.

Pony.ai is a leading global autonomous driving technology company founded in 2016, focused on bringing safe, sustainable, and accessible autonomous mobility to the world. The company develops a full-stack autonomous driving technology, leveraging its core "virtual driver" system. Pony.ai has accumulated millions of kilometers in autonomous road testing in complex scenarios, including challenging weather and road conditions, and has secured licenses to test and operate autonomous vehicles globally. Its business units include Robotaxi for everyday travel, Robotruck for commercial logistics, and solutions for Personally Owned Vehicles (POV), aiming to deliver superb autonomous driving solutions across various industries and markets.

bili-hardcore is an AI-powered tool designed to automate the process of answering questions for Bilibili's hardcore member exams. Unlike OCR-based solutions, it directly interacts with the Bilibili API, ensuring higher accuracy and efficiency. The tool supports various large language models, including DeepSeek (V3.1) and Gemini (gemini-2.5-flash), with options for custom OpenAI-style APIs like those from Volcengine and SiliconFlow. Users can configure their preferred model and API key, and the tool handles the login via QR code and automatic question answering. It's crucial for users to have a Bilibili account at level 6 or above to participate in the hardcore member trials. The tool also provides guidance on troubleshooting common issues like QR code display problems, low accuracy, or API errors, and emphasizes responsible use in compliance with Bilibili's rules.

Stellon Labs is an AI research lab dedicated to developing powerful, tiny AI models specifically optimized for edge applications. Their focus is on creating 'frontier AI' solutions that can operate efficiently on minimal hardware, making advanced artificial intelligence accessible for devices with limited computational resources. The lab aims to push the boundaries of AI performance in constrained environments, enabling new possibilities for on-device intelligence without requiring extensive infrastructure. Their work is geared towards practical applications where low-power and small-footprint AI is crucial.

AskVideo.ai is a free tool designed to transform how users interact with YouTube videos. By simply pasting a YouTube URL, the AI processes the video content, allowing users to ask questions and receive instant, accurate answers with corresponding timestamps. This eliminates the need for manual scrubbing through long videos, making learning and information extraction significantly more efficient. It's ideal for academic learning, tutorials, business insights, research, and team collaboration, offering features like AI-powered Q&A, video transcripts, and even a command-line interface for developers. The platform aims to save users hours of time by providing direct access to specific information within video content.

Tap Mobile is a company focused on developing AI-powered utility applications that are used by over 400 million people across 130+ countries. Their product portfolio includes a range of tools designed to make technology more accessible, such as document scanning and photo enhancement apps. While the specific app "Luna - Your Ai Companion" is not detailed on the provided website, Tap Mobile's overarching mission is to integrate artificial intelligence into daily mobile experiences, providing practical and innovative solutions for a broad global audience. Their apps aim to simplify tasks and enhance user interaction through advanced AI capabilities.

HiveChat is an AI chat application specifically developed for small and medium-sized teams, offering support for a wide range of AI models including Deepseek, OpenAI, Claude, and Gemini. It provides robust administrative features, enabling a single administrator to configure and manage AI model access for the entire team. Key functionalities include user grouping, setting different available models per group, and assigning monthly token limits. The platform also supports various login methods such as email, WeChat Work, DingTalk, and Feishu, ensuring flexible integration into existing team workflows. HiveChat facilitates cloud data storage and features like DeepSeek thought chain display, LaTeX and Markdown rendering, and image understanding.

my-neuro is an open-source project designed to help users create their own personalized AI desktop companions. Inspired by Neuro Sama, this tool allows for extensive customization of characters, including voice, personality, and appearance, compatible with various Live2D models. It boasts ultra-low latency responses, with conversations responding in under one second, and supports both local inference with open-source LLMs and integration with closed-source AI models via DMXAPI. Key features include long-term memory, visual recognition, voice cloning, and LLM training, enabling the AI to remember user interactions, understand visual cues, and adapt its responses. The project also plans to integrate advanced human-like interaction designs, such as real-time interruptions, emotional responses, and desktop control capabilities, making it a versatile platform for building deeply personal AI companions.

Pytorch-bert-crf-ner offers a PyTorch implementation for Korean Named Entity Recognition (NER) tagging, leveraging the power of BERT and CRF models. This open-source tool is specifically designed to assist in Korean Natural Language Processing (NLP) tasks and research. It provides functionalities to identify and classify named entities such as persons, locations, organizations, dates, and more within Korean text. The repository includes examples, data utilities, and training scripts, making it suitable for developers and researchers working with Korean language data who need to implement or experiment with NER models.

Pytorch-maddpg offers a PyTorch implementation of the Multi-Agent Deep Deterministic Policy Gradient (MADDPG) algorithm, a key approach in multi-agent reinforcement learning. This open-source project is hosted on GitHub and is designed for researchers and developers working on complex multi-agent systems. The implementation includes a modified Waterworld environment, where agents (evaders, pursuers, poisons) interact under specific physical rules, allowing for experimentation with cooperative behaviors. It supports features like agents bouncing off walls and requiring exact cooperation for rewards, making it a valuable tool for studying multi-agent coordination and policy learning.