AI Agents & Automation

Haechi AI offers free, AI-powered fraud protection specifically designed for elderly Americans and their families. The platform screens incoming phone calls in real-time, detecting spoofed numbers, impersonation attempts, and high-pressure tactics before a user answers. It also allows users to photograph suspicious physical mail for instant analysis, identifying lottery scams, fake government notices, and fraudulent checks. Haechi AI provides ongoing fraud education through weekly briefings and real-time alerts, empowering users to recognize new scam techniques. A family dashboard feature enables adult children to monitor a parent's protection status, review flagged threats, and receive notifications, offering peace of mind. The service emphasizes data privacy with 256-bit encryption and a strict no-data-selling policy.

BMInf (Big Model Inference) is an open-source toolkit designed to facilitate efficient inference for large-scale pretrained language models (PLMs). It enables the execution of models with over 10 billion parameters, even on low-resource hardware like a single NVIDIA GTX 1060 GPU. The tool offers significant performance improvements over existing PyTorch implementations, particularly for GPUs like V100 or A100. BMInf 2.0.0 introduced compatibility with any transformer-based model, making it a versatile solution for researchers and developers working with big AI models. It provides methods for automatic model conversion using `bminf.wrapper` or manual replacement of modules like `torch.nn.ModuleList` and `torch.nn.Linear` for optimized performance.

ROK Solution is a comprehensive hyperautomation platform designed to revolutionize business processes by combining workflow, BPM, RPA, AI, and no-code capabilities. It enables organizations to automate processes without extensive coding, accelerating application creation and streamlining operations. The platform emphasizes balancing agility with governance and cybersecurity, offering features like automated org chart generation, Identity Governance & Administration (IGA) for secure access, and built-in Generative AI for application creation. ROK Solution aims to reduce operational costs, improve customer satisfaction, and enhance security, making it suitable for digital transformation in both business and governmental organizations that require stringent security and compliance.

LitterBox provides a controlled sandbox environment specifically designed for security professionals, including malware developers and red teamers. It enables users to test evasion techniques against modern detection mechanisms, validate detection signatures before deployment, and analyze malware behavior in an isolated environment. The platform ensures payloads remain in-house, preventing exposure to external security vendors and confirming functionality without triggering production security controls. LitterBox features LLM-assisted analysis capabilities through the LitterBoxMCP server, offering advanced analytical insights using natural language processing. It supports both static and dynamic analysis, including file identification, entropy analysis, executable analysis for PE files, document analysis, and LNK analysis. Dynamic analysis offers runtime behavioral monitoring, memory inspection, and detection of techniques like process hollowing and code injection. The tool also includes HolyGrail for BYOVD analysis and Blender/FuzzyHash modules for code similarity and process comparison.

xlstm is the official GitHub repository for the xLSTM, a new Recurrent Neural Network architecture based on the original LSTM. This tool provides the necessary code and resources for researchers and practitioners to implement and experiment with this novel LSTM variant. xLSTM utilizes Exponential Gating with appropriate normalization and stabilization techniques, along with a new Matrix Memory, to overcome the limitations of traditional LSTMs. It demonstrates promising performance in Language Modeling compared to Transformers or State Space Models. The repository includes a 7B parameter xLSTM Language Model trained on 2.3T tokens, optimized for fast and efficient inference, and offers detailed instructions for installation and usage.

XManager is an open-source platform developed by Google DeepMind designed for managing machine learning experiments. It simplifies the process of packaging, running, and tracking ML experiments, whether executed locally or on Google Cloud Platform (GCP). The platform offers Python APIs that allow users to interact with experiments through launch scripts, providing a structured approach to ML development. Key features include defining executable specifications for binaries, containers, and Python modules, as well as executor specifications for running jobs on various platforms like local machines, Vertex AI, or Kubernetes. XManager supports both single jobs and JobGroups for gang scheduling, making it suitable for complex, multi-component experiments. It also facilitates the management of hyperparameters and resource requirements for each job.

Flashback empowers businesses to build transformative applications faster and smarter through a unified multi-cloud platform. It offers an Enterprise AI Gateway for secure access and policy control for AI usage, supporting OpenAI, Anthropic AI, AWS Bedrock, and Google VertexAI. This gateway enables pre-request policy enforcement, data anonymization, compliance controls, and cost visibility. Additionally, Flashback provides a Multi-Cloud Storage Gateway to empower storage with a multi-cloud strategy, allowing users to connect public and private S3-compatible clouds, optimize storage costs, and manage redundancy and recovery from a single control panel. The platform emphasizes governance, privacy, and observability for cloud operations.

xiaozhi-esphome provides alternative code to use Xiaozhi AI devices as voice assistant satellites for Home Assistant, leveraging ESPHome. This open-source project simplifies the integration of compact Xiaozhi-based devices into a smart home setup, allowing them to act as voice assistants. The project offers a quick start guide for installation, including steps for connecting devices via USB, configuring them with ESPHome Web, and integrating them into Home Assistant. It supports a growing list of devices like Espressif EchoEar, Spotpear Ball, Muma Box, and various Waveshare and Guition models. The repository also includes links to purchase supported devices and 3D print files for accessories.

Trinity-RFT is a comprehensive, open-source framework designed for the reinforcement fine-tuning (RFT) of large language models (LLM). It offers a general-purpose, flexible, and scalable solution by decoupling the RFT process into three core components: Explorer for generating experience data, Trainer for updating model weights, and Buffer for managing data pipelines. This architecture supports various RFT modes, including synchronous/asynchronous, on-policy/off-policy, and online/offline RL, allowing for independent scaling of rollout and training. Trinity-RFT also provides robust support for agentic RL workflows, full-lifecycle data pipelines with active data management, and a user-friendly design with plug-and-play modules and graphical interfaces. It supports a wide array of algorithms like PPO, GRPO, DPO, and CHORD.

AirLLM is an open-source framework designed to optimize inference memory usage for large language models, enabling 70B models to run on a single 4GB GPU without requiring quantization, distillation, or pruning. It also supports running 405B Llama3.1 models on 8GB VRAM. The tool offers features like model compression for up to 3x inference speed improvement, support for various LLMs including Llama, Qwen, ChatGLM, Baichuan, Mistral, and InternLM, and compatibility with MacOS. AirLLM simplifies the inference process with an AutoModel feature that automatically detects model types and provides prefetching to overlap model loading and computation for enhanced speed.

Mirai is an AI platform designed to convert, optimize, distribute, and run AI models with the fastest inference engine on Apple Silicon. It allows developers to deploy models on Mac, iPhone, and iPad, ensuring offline and private execution. Mirai offers one-line model conversion, quantization with high quality, and supports various architectures. It leverages the full potential of Apple Silicon's Neural Engine and unified memory bandwidth for real-time generation on devices. The platform supports use cases like text summarization, classification, routing, and translation, with upcoming voice features. Mirai also enables seamless distribution, zero inference cost, and keeps data on the device, making it ideal for applications requiring privacy and low latency.

AudioCraft is a comprehensive PyTorch library designed for deep learning research in audio generation. It provides both inference and training code for advanced AI generative models, including MusicGen for controllable text-to-music generation and AudioGen for text-to-sound. The library also integrates the state-of-the-art EnCodec audio compressor/tokenizer, Multi Band Diffusion for EnCodec-compatible decoding, and MAGNeT for non-autoregressive text-to-music/sound. Additionally, it offers AudioSeal for audio watermarking and JASCO for high-quality text-to-music conditioned on chords, melodies, and drum tracks, making it a powerful toolkit for researchers and developers in the audio AI domain.

Luel is a two-sided marketplace designed to facilitate the exchange of high-quality AI training data, specifically focusing on video and audio content. It connects AI development teams seeking specific datasets with contributors who can provide the necessary video, audio, and image content. The platform ensures that all training data is curated, rights-cleared, and verified, making it suitable for commercial use. Enterprises can access a catalog of premium datasets or request custom data collection campaigns, benefiting from enterprise-grade quality and compliance. Contributors, on the other hand, can upload their content, such as cooking tutorials, voice samples, or product photos, to earn income with fast payouts and fair rates, without upfront costs.

OmAgent is a Python library designed to simplify the development of multimodal language agents. It abstracts away complex engineering challenges such as worker orchestration, task queues, and node optimization, providing a straightforward interface for defining agents. The library supports various multimodal interactions, including native integration with VLM models, real-time APIs, computer vision models, and mobile device connections. This enables developers and researchers to build agents that can process and reason over diverse inputs like text, images, video, and audio. OmAgent also offers a flexible agent architecture with a graph-based workflow orchestration engine and multiple memory types for contextual reasoning. It includes state-of-the-art unimodal and multimodal agent algorithms like ReAct, CoT, and SC-CoT, and supports local model deployment via Ollama or LocalAI, alongside a fully distributed architecture with custom scaling options.

Waveye specializes in AI-driven imaging radars, delivering ultra high-resolution Lightweight Imaging Radar (LIR) technology with deeply-integrated radar AI. This advanced perception system is designed to enable robust autonomy at scale across multiple industries. Key performance indicators include a native angular resolution of 0.5 / 0.9 in azimuth and elevation, a wide 160-degree field of view in azimuth and 40 degrees in elevation, and an operating range exceeding 200 meters. The technology is capable of over 5000 detections in typical urban scenes, making it suitable for demanding applications. Waveye's solutions are particularly relevant for off-road autonomy, robotics, and automotive sectors, providing enhanced object detection and environmental understanding.

PromptLayer is a robust AI development tool designed for prompt engineers and developers working with large language models. It functions as middleware, seamlessly integrating with the OpenAI Python library to log and manage all API requests and prompts. Users can track, debug, and replay past completions, offering a comprehensive solution for prompt versioning, testing, and monitoring. The library provides convenient access to the PromptLayer API, allowing for prompt template retrieval, listing, publishing, and cache invalidation. It also includes features for manual request logging, request annotation with metadata, prompt linkage, scores, and groups, and even a decorator for tracing custom functions. With support for both synchronous and asynchronous operations, PromptLayer streamlines the development workflow for AI applications.

Wonin AI is a global leader in providing automated surveillance solutions, video content analysis, and Intelligent Video Analytics. Their analytics technology covers a wide range of applications, including facial recognition, retail business intelligence, and advanced security analytics. Wonin AI's offerings include AI-based video analytics for debris and garbage detection, no helmet detection, speed detection, stopped vehicle detection, automatic number plate recognition, human pattern recognition, video-based fire detection, armed person detection, camera tampering detection, object removal detection, and wrong way detection. They also provide central monitoring systems, cloud surveillance, and command control centers. Wonin AI has pioneered the use of AI in various industries such such as Oil & Gas, Smart Cities, Defence, Metro, Airport, Hospitality, Healthcare, IT Solutions, Smart City/City Surveillance, Public Services, Retail, and Transport Storage.

whisper-vits-svc is an open-source core engine for singing voice conversion and singing voice cloning, built upon the VITS framework. It leverages variational inference with adversarial learning for end-to-end voice transformation. Designed for deep learning beginners, the project requires basic knowledge of Python and PyTorch. Key features include support for multiple speakers, the ability to create unique speakers through mixing, and conversion of voices even with light accompaniment. Users can also edit F0 using Excel and benefit from various model properties like strong noise immunity and improved conversion stability. The tool does not support real-time voice converting and focuses on practical application for learning deep learning concepts.

Koxy AI is a no-code platform designed for building AI-powered serverless backends without writing any code. It provides a globally distributed infrastructure, ensuring fast and secure solutions delivered at the edge. The platform is tailored for front-end developers, enabling them to create modern backends effortlessly. Koxy AI emphasizes ease of use, allowing users to develop robust serverless applications with no limits on scalability or functionality. This tool simplifies backend development, making advanced AI capabilities accessible to a broader range of developers.

QARL AI specializes in AI-powered learning, focusing on the deployment of video avatars to facilitate experiential learning. The company is currently operating in stealth mode, indicating that specific functionalities and detailed product offerings are not yet publicly disclosed. However, its core mission revolves around enhancing learning experiences through advanced AI technology, particularly through interactive video avatars. Interested parties are encouraged to reach out directly to learn more about their solutions and capabilities.

Euno is an AI context platform designed for enterprise data, transforming metadata into automated and trusted context for AI agents. It enables AI agents to act reliably and safely at enterprise scale by providing them with everything they need to know about core data. Euno connects to AI agents to ensure they query the right data using the correct logic, even in complex environments. Key features include real-time context graph construction with lineage, activity, health, and business logic, built-in governance automation, and automated labeling of assets based on custom rules. It helps organizations avoid common AI failures like hallucinations and inconsistencies by grounding AI decisions in accurate, contextualized data.

labml is an open-source tool designed to monitor deep learning model training and hardware usage, accessible from both mobile phones and laptops. It offers easy integration with just two lines of code, allowing users to track experiments, including git commits, configurations, and hyperparameters. The tool also provides real-time monitoring of hardware usage on any computer. Key features include an API for custom visualizations, pretty logs of training progress, and compatibility with frameworks like PyTorch and TensorFlow. Users can host their own experiment server and access the user interface locally or on a separate machine, making it a flexible solution for deep learning practitioners.

GNOSS is a Semantic AI platform that leverages Neurosymbolic AI on knowledge graphs to automate critical decision-making. It transforms complex, fragmented data into actionable and auditable knowledge, primarily for government and business organizations. The platform offers solutions like Semantic AI Platform for unifying information, Odysseus for deploying and monitoring robust AI, and GNOSS Cognitive Services & AI Platform for integrating reasoning, learning, and explainability. GNOSS emphasizes traceable, explainable, and auditable deterministic reasoning, ensuring transparency and control in critical applications across various industries including defense, public administration, logistics, education, legal, culture, healthcare, finance, retail, and tourism.

NVIDIA Run:ai is an enterprise platform designed to accelerate AI and machine learning workflows by addressing key infrastructure challenges. It provides dynamic resource allocation, comprehensive AI lifecycle support, and strategic resource management to enhance GPU efficiency and workload capacity. The platform supports public clouds, private clouds, hybrid environments, and on-premises data centers, offering unparalleled flexibility. NVIDIA Run:ai centralizes and automates AI workload execution, ensuring optimal utilization and alignment with business objectives. It integrates seamlessly with major AI frameworks and machine learning tools, reducing operational costs and accelerating AI initiatives from development to deployment.