AI Agents & Automation

OpenAI-Unity is an unofficial Unity package designed to help game developers seamlessly integrate the OpenAI API into their Unity projects. This package enables the direct use of OpenAI's powerful AI capabilities, such as ChatGPT for conversational AI and DALL.E for image generation, within the Unity game engine. Developers can make asynchronous requests to the OpenAI API, including stream requests for real-time interactions. The package provides clear instructions for importing, setting up an OpenAI account, and securely saving API credentials. It also includes sample projects like a ChatGPT-like chat example and a DALL.E text-to-image generation example, making it easier for developers to get started with AI-driven game features. It supports various Unity versions for different platforms, including WebGL builds, though some known issues with WebGL are documented.

rag_api is an ID-based Retrieval-Augmented Generation (RAG) FastAPI designed for asynchronous and scalable document indexing and retrieval. It seamlessly integrates Langchain with PostgreSQL/pgvector, allowing for efficient management of documents organized into embeddings by file_id. This approach enables targeted queries when combined with file metadata, making it ideal for applications like LibreChat or any other ID-based RAG use case. The API offers robust document management features, including methods for adding, retrieving, and deleting documents, alongside asynchronous support for enhanced performance. It supports various embedding providers such as OpenAI, Azure, Hugging Face, Google GenAI, VertexAI, and Ollama, and allows for configurable chunking, batch processing, and distance thresholds to optimize retrieval quality and cost.

Spark NLP is a state-of-the-art Natural Language Processing library built on top of Apache Spark, designed for machine learning pipelines that require scalability in distributed environments. It offers a comprehensive suite of NLP tasks including Tokenization, Part-of-Speech Tagging, Named Entity Recognition, Sentiment Analysis, Machine Translation, and Question Answering. The library supports over 100,000 pretrained pipelines and models across more than 200 languages, integrating seamlessly with modern transformer models like BERT, Llama-2, and GPT2. Spark NLP also provides easy model importing from frameworks such as TensorFlow, ONNX, OpenVINO, and Llama.cpp, enhancing flexibility for developers working with diverse machine learning ecosystems. It supports Python, Scala, Java, and Kotlin, and is compatible with platforms like Databricks, EMR, and Google Cloud Dataproc.

thepipe is a powerful Python package designed to extract clean, structured, and multimodal data from a wide array of complex documents. Leveraging vision-language models (VLMs), it excels at scraping markdown, tables, images, text, video, and audio from sources including PDFs, URLs, Word documents, PowerPoints, Python notebooks, and even GitHub repositories. It offers AI-native file-type detection, layout analysis, and structured data extraction, working seamlessly with any LLM, VLM, or vector database. The tool provides various chunking methods to manage token limits and integrates with OpenAI and LlamaIndex, making it ideal for RAG frameworks and advanced data processing workflows.

TangoFlux is an advanced text-to-audio generation tool developed by declare-lab, accepted to ICLR 2026. It leverages FluxTransformer blocks, including Diffusion Transformers (DiT) and Multimodal Diffusion Transformers (MMDiT), conditioned on textual prompts and duration embeddings. The tool is capable of generating high-fidelity 44.1kHz stereo audio, up to 30 seconds in length, with remarkable speed, achieving generation in about 3 seconds on a single A40 GPU. TangoFlux learns a rectified flow trajectory to an audio latent representation encoded by a variational autoencoder (VAE). Its training pipeline involves pre-training, fine-tuning, and preference optimization using CRPO (Clap-Ranked Preference Optimization) for flow matching. The tool offers various interfaces including a Python API, CLI, and integration with ComfyUI, making it accessible for researchers and developers.

Cycls is a comprehensive platform designed for building, deploying, and utilizing AI agents. It enables teams, developers, and businesses to put intelligent agents to work rapidly. The platform features the Cycls Agent SDK, allowing users to code agents with any AI tool and ship them in minutes, complete with instant rich UI and fully typed REST API without requiring zero configuration. Cycls Cloud provides product infrastructure for AI, scaling from zero to production volume with built-in authentication, security, and analytics. It supports agents that can reason, remember conversations, connect to tools, and take real action. The platform is framework and LLM agnostic, built for production with secure sandboxing, streaming API, built-in authentication, native UI components, and observability features.

Covariant specializes in developing and deploying Robotics Foundation Models (RFM-1) to power advanced automation solutions for the logistics and retail sectors. Their core product, the Covariant Brain, is an AI robotics platform trained on a vast multimodal robotics dataset from warehouses globally. This allows robots to pick virtually any SKU or item from day one, addressing fluctuating demand and labor gaps in warehouse operations. The platform facilitates fleet learning, enabling robots across a network to leverage shared insights and adapt to dynamic business needs. Covariant aims to bring AI from research into practical, reliable, and flexible real-world applications, ensuring performance for leading retailers and logistics providers.

Temporal AI Agent is a demo showcasing a multi-turn conversation with an AI agent operating within a Temporal workflow. This agent is designed to collect information to achieve a specific goal, utilizing various tools in the process. It supports both native tools, implemented directly in the codebase, and external tools accessed via the Model Context Protocol (MCP), allowing integration with services like Stripe or databases. The system offers single-agent mode by default and an experimental multi-agent/multi-goal mode, enabling users to switch between different agent types during conversations. Goals are organized by category and can leverage both types of tools, with the AI clarifying and requesting missing information as needed. It can be configured to use various LLMs supported by LiteLLM, including OpenAI, Anthropic Claude, Google Gemini, and Ollama models.

txtai is an open-source AI framework designed for semantic search and LLM orchestration, providing an embeddings database that integrates vector indexes, graph networks, and relational databases. This unification enables efficient vector search and comprehensive knowledge management within language model workflows. The tool is hosted on GitHub, suggesting a developer-centric approach and offering various features for code creation, developer workflows, and application security. It supports automation with GitHub Actions, package hosting, and project management features, making it suitable for teams looking to build and deploy intelligent applications.

Sony AI is a dedicated research and development organization established by Sony to push the boundaries of artificial intelligence. Their work spans multiple critical areas such as Gaming & Interactive Agents, Imaging & Sensing, Robotics, Scientific Discovery, AI for Creators, and AI Ethics. They collaborate with various Sony entities and external partners to develop AI solutions that enhance human creativity, solve complex global challenges, and contribute to scientific progress. Key projects include Gran Turismo Sophy, an advanced AI agent for racing simulation, and the development of ethical AI frameworks and datasets like FHIBE for fair human-centric image benchmarking. Sony AI focuses on both theoretical advancements and practical applications of AI.

TrustLLM is a European initiative dedicated to democratizing trustworthy and factual large language model (LLM) technology. Funded by the European Union, the project aims to develop LLMs that are reliable and accurate, with a particular focus on covering underrepresented languages within Europe. The project, which commenced in November 2023 and is set to run for three years, actively contributes to the broader European AI ecosystem. TrustLLM engages in research, hosts webinars, and participates in events to foster collaboration and disseminate knowledge on responsible LLM deployment. Key figures like Fredrik Heintz from Linköping University lead the project, emphasizing the importance of collaboration and autonomy in shaping AI's future in Europe.

PenguinBot AI is a powerful platform designed for teams to deploy AI agents to Azure effortlessly. It simplifies the process of launching secure, multi-channel AI assistants, ensuring fast onboarding and seamless integration into existing workflows. The tool focuses on providing an agentic AI experience, acting as an always-on AI teammate that can manage various tasks. Built for production workflows, PenguinBot AI supports secure deployments and is ideal for organizations looking to leverage AI agents within their Azure environment for enhanced productivity and automation.

The House Of AI specializes in helping businesses integrate sustainable growth with impactful AI solutions. Their ethos centers on trustable AI, enabling ethical and governed business practices through regulatory frameworks, transparent AI technologies, and human-in-the-loop approaches. They offer tailored learning paths and workshops to bring teams up to speed with AI tools and techniques, alongside strategic services like complexity assessment and technology selection. Their intervention areas span from building foundations with generative AI and synthetic data, to optimizing operations with LLM service design and causal AI, and disrupting markets with Human-AI Interaction and AI Agents. They also focus on business automation across sales, brand engagement, operations, and customer care.

Naptha AI offers a modular platform designed for autonomous agents, empowering developers to create AI applications, perform AI research, and scale cooperative AI systems. The platform addresses common challenges in AI agent interaction, aiming to improve user experience and reduce churn. Key solutions include product-led features like in-editor agent-guided walkthroughs via WizardKit.ai, simplified agent hosting with MCP and A2A servers (AutoMCP and AutoA2A), and secure agent authentication for permissioned data sources through OneDollarOAuth.com. Additionally, it provides benchmarking tools like StackBench.ai to ensure effective use of specific libraries, helping businesses leverage agentic software development.

Innoneur is an R&D-focused AI software company founded in 2014, providing bespoke AI/ML solutions, multi-agent systems, and enterprise transformation services. They leverage their research expertise to build AI-first systems from the ground up, emphasizing ethical performance and sustainable growth. Innoneur offers expert guidance on responsible AI strategies, prioritizing integrity, inclusivity, and compliance to ensure projects deliver meaningful impact. Their dedicated consultants provide ongoing support and advice throughout each project phase, aiming to achieve AI-driven results. Notable success stories include LiMore.ai for multi-agent customer support, AI-driven recruitment automation, and agentic real-time analytics for finance.

Inquizyt Socle is a sovereign platform designed for specialized AI agents that automate end-to-end business processes. Deployable on-premise, it guarantees data sovereignty, banking-level security, and compliance with GDPR and AI Act. The platform's intelligent agents understand specific business contexts and rules, significantly reducing processing times and operational costs. It offers various specialized agents for tasks like document intelligence, process automation, data analytics, and compliance monitoring. Inquizyt aims to boost productivity by 300% and achieve ROI within 3-6 months, making advanced AI accessible for businesses without requiring extensive technical expertise.

linkhub.ai offers Agentic AI solutions designed to automate and transform enterprise operations. The platform develops autonomous Agentic Systems capable of reasoning, planning, and executing complex business tasks end-to-end. Key offerings include AI agents for customer service, sales, and multi-purpose agentic systems for back-office and critical operations. Additionally, linkhub.ai provides AI evaluators to audit other AI projects for hallucination, quality, and precision using advanced GenAI metrics. They also offer AI Transformation consulting, guiding organizations from initial diagnosis to implementation and scaling of AI projects, ensuring tangible business impact and secure integration with existing systems.

Valyu offers an AI Search API designed for AI agents and knowledge work, providing real-time structured content from both web and proprietary sources. It allows users to perform deep research tasks, powering applications where accuracy and depth are critical. The platform connects AI agents to the web and beyond, offering real-time search, extraction, and deep research through a single API. Valyu supports various data sources, including web content, proprietary research, academic papers, and financial data, all queryable in natural language. It is built for enterprise workloads with SOC2 compliance, zero data retention, SSO, and high capacity limits, making it suitable for startups and large organizations alike.

Spheron Network is an enterprise-grade marketplace for on-demand GPU rental, specializing in high-performance NVIDIA GPUs like H100, A100, and B200. It provides instant deployment from Tier 3/4 data centers, eliminating the need for long-term contracts and offering flexible pay-per-hour pricing. The platform supports various AI workloads, including AI training, LLM inference, machine learning, and deep learning. Users can access a multi-provider GPU selection through a single unified interface, with support for SXM5 InfiniBand and PCIe configurations. Spheron aims to make powerful compute resources accessible and scalable for developers, AI researchers, and enterprises.

Vue.ai by M2P is an Enterprise AI Orchestration Platform designed to accelerate AI transformation for businesses. It offers a composable, modular, and AI-native architecture that allows for fast onboarding and quick time to value, avoiding lengthy data transformation cycles. The platform automatically cleans and enriches data, making it relevant across various use cases. Vue.ai focuses on delivering business outcomes with a 30:60:90 promise, ensuring pilot go-live in 30 days, ROI proof in 60 days, and scalability in 90 days. It provides solutions across Data Hub for data cleanup and product tagging, Customer Hub for personalization and audience building, Automation Hub for intelligent document processing and workflow automation, and Optimization Hub for sales efficiency and lead generation. This platform is ideal for industries like eCommerce, retail, financial services, insurance, logistics, and healthcare.

AZTRA provides cutting-edge AI and machine learning solutions designed for enterprise businesses. The platform focuses on autonomous decision-making, advanced analytics, decision intelligence, and data & AI engineering to transform operations. It ingests and standardizes real-time data, creating a unified foundation for analysis, and builds custom AI models tailored to specific business goals. AZTRA translates insights into operational KPIs and delivers intelligence directly within existing workflows, enabling faster, more informed decisions. The company specializes in building scalable, secure, and compliant agentic AI systems that automate routine tasks and elevate critical operations, helping teams move faster with clarity and measurable impact.

Celestial Dynamics is a leading technology laboratory and NVIDIA Solution Partner specializing in developing and implementing enterprise-grade artificial intelligence solutions. They focus on AI Factory and AI Enterprise, transforming businesses with measurable results and proven ROI. Their services include strategic consulting to reduce risk and align executive vision, custom AI system development and integration (from use case to MVP in production), and automated solutions via pre-built Neural Engines like MATRIA AI and TalentGraph AI. They leverage NVIDIA technologies such as RAPIDS, NeMo, Riva, TAO Toolkit, and Omniverse to ensure high performance and reliability for their clients.

Pacific Data Integrators (PDI) provides cutting-edge AI solutions and services, focusing on data management, cloud modernization, and advanced analytics. Their offerings include a Generative AI Platform for creating new experiences, Data Science in a Box for democratizing data science, and Retail-AI for personalized product recommendations and supply chain analytics. PDI also specializes in semantic search engines and legacy code modernization using Gen AI. They offer end-to-end services, including advanced LLMs and Data Science expertise, with a simplified and transparent workflow. Their proficiency spans major platforms like Informatica, AWS, Google Cloud, Azure, Databricks, and Snowflake, ensuring scalable, secure, and cost-effective solutions for various industries.

Neuromation specializes in enterprise AI transformation and MLOps services, offering end-to-end AI implementation for businesses. Their services range from adoption advisory to custom infrastructure setup, supported by PhD-level ML researchers and full-stack engineers. Neuromation helps organizations overcome the challenges of bringing machine learning and deep learning projects into production by providing managed MLOps platforms. This approach aims to reduce the time ML teams spend on infrastructure setup and maintenance, allowing them to focus on experimentation and innovation. They also offer specific AI solutions, such as a highly accurate AI-based face mask detection system, which is easy to deploy and compatible with various camera systems.