Research & Education

ru-gpts is a collection of open-source Russian GPT-3 models, including ruGPT3XL, ruGPT3Large, ruGPT3Medium, and ruGPT3Small, alongside a ruGPT2Large model. These autoregressive transformer language models are trained on an extensive dataset of the Russian language, offering different sequence lengths and attention mechanisms. The repository provides detailed setup instructions for both Colab and local environments, usage examples for generation, and finetuning capabilities. Pretraining details for each model, including dataset size, training time, and hardware used, are also available. The project is developed by AI-Forever and Sberbank-AI, making advanced Russian language models accessible for various applications.

CiCi Chat Bot is an AI-powered careers assistant from CareerChat UK, designed to help young people and adults navigate complex career decisions. It offers expert advice, integrates with Careers Hubs, and provides personalized, LMI-driven guidance that meets statutory careers requirements. The platform also includes CiCiChat webinars for prompt engineering, research into AI trends, and consultancy services. CiCi can be white-labelled, customized with conversation flows, and integrated with existing data sets like course directories. It also offers bespoke services such as evidence-based intelligence dashboards and MIS integration, aiming to transform how organizations deliver scalable and compliant careers support.

Cogram is an AI-native platform specifically designed for architects and engineers, offering a suite of tools to enhance project documentation, collaboration, and design processes. It addresses the challenges of outdated software by providing AI-powered solutions for tasks such as RFI and submittal management, automated meeting minutes, email filing, and field reports. The platform integrates with existing AEC tools like Procore, Deltek, Unanet, Microsoft Teams, Outlook, Autodesk, Zoom, and Google Meet, ensuring compatibility with current workflows. Cogram emphasizes data security with SOC 2 Type II certification and offers flexible deployment options, including hybrid cloud. It aims to improve efficiency by automating repetitive administrative tasks, saving users several hours per week.

Elicit is an AI-powered research assistant designed to significantly enhance the efficiency and evidence-based nature of scientific research. It allows users to search over 138 million academic papers and 545,000 clinical trials using semantic search, eliminating the need for precise keywords. The tool can generate high-quality research reports, customizable to specific papers and information, and supports systematic literature reviews by automating screening and data extraction, potentially saving researchers up to 80% of their time. Elicit also offers features like a personal library for organizing sources, alerts for new research, and an API for integrating its search and report generation capabilities. It emphasizes accuracy, transparency with sentence-level citations, and goes beyond simple chat to provide interactive tables and multi-step workflows.

Study Materials is a comprehensive online repository designed to support B.Tech (Hons) CSE students with their academic journey. The platform offers an extensive collection of semester-wise study materials, including detailed notes, lab work, and additional resources for various engineering subjects. Users can find content spanning from Engineering Mathematics and Data Structures to Artificial Intelligence, Machine Learning, and Cloud Computing. The website is organized for easy navigation, allowing students to access specific semester content, view syllabi, and explore questionnaires. It also includes resources for commonly used commands and showcases various projects, making it a valuable academic support and learning platform.

large-language-models is an open-source repository from Databricks Academy, providing comprehensive notebooks and slides for a Large Language Models (LLMs) Specialization course. This educational resource is designed for individuals looking to understand the application and production deployment of LLMs, available on edX and Databricks Academy. The repository includes practical notebooks covering topics such as LLM applications, embeddings, vector databases, multi-stage reasoning, fine-tuning, evaluation, societal impacts, and LLMOps. Users can easily import the repository into Databricks workspaces, with clear instructions for setting up Git credentials and importing files. It also specifies recommended Databricks Runtime versions and cluster settings, including GPU requirements for fine-tuning notebooks, ensuring an optimized learning environment.

Kaggle serves as the world's AI proving ground, bringing together millions of builders, researchers, and labs to discover what truly works in artificial intelligence. The platform facilitates the evaluation of AI agents, models, and frontier technology through crowdsourced benchmarks, competitions, and hackathons. It provides a robust environment for data scientists, machine learning engineers, and AI enthusiasts to collaborate, learn, and showcase their skills. Kaggle is a central hub for community-led innovation, offering resources and tools to advance the field of AI.

VideoGPT is an open-source project for video generation, leveraging VQ-VAE (Vector Quantized Variational AutoEncoder) and Transformer architectures. It learns downsampled discrete latent representations of raw video using 3D convolutions and axial self-attention. A GPT-like architecture then autoregressively models these discrete latents with spatio-temporal position encodings. The tool is designed for researchers and developers interested in AI video creation, offering a reproducible reference for transformer-based video generation models. It can generate samples competitive with state-of-the-art GAN models on datasets like BAIR Robot, UCF-101, and TGIF. The project includes scripts for training VQ-VAE and VideoGPT models, as well as for sampling and evaluation.

Indonesia AI (PT. Teknologi Artifisial Indonesia) is dedicated to fostering digital transformation in Indonesia through Artificial Intelligence. The platform provides comprehensive AI learning resources, including online courses via Zoom or Learning Management System (LMS), covering topics from Fundamental AI to Generative AI and AI for Kids. Beyond education, Indonesia AI offers AI consulting services, conducting research and development for AI solutions across various sectors like finance, healthcare, and manufacturing. They also provide AI corporate training, focusing on practical applications and post-training support to integrate AI effectively into businesses. Key services include anomaly and fraud detection, image classification, object detection and segmentation, face recognition, object counting, and video tracking.

llm-zoomcamp is a comprehensive, free online course designed to equip participants with the skills to build AI systems using Large Language Models (LLMs). Over 10 weeks, the curriculum covers essential topics such as Retrieval-Augmented Generation (RAG), vector search, evaluation methodologies, and monitoring techniques. The course provides hands-on experience with tools like OpenAI API, Elasticsearch, and Qdrant, enabling learners to index and retrieve data efficiently. It also includes bonus modules on agents and an end-to-end project to build a fitness assistant, culminating in a capstone project where participants apply all learned concepts. The course is self-paced, offering flexibility for learners to watch videos, complete homework, and share projects for feedback.

UniDiffuser is an open-source framework offering code and models for multi-modal diffusion research, based on the paper "One Transformer Fits All Distributions in Multi-Modal Diffusion." It unifies learning for marginal, conditional, and joint distributions by predicting noise in perturbed data across different modalities. The tool uses a transformer architecture to handle various input types and can perform image, text, text-to-image, image-to-text, and image-text pair generation. It supports tasks like image variation and text variation, and can be integrated with the Hugging Face Diffusers library for ease of use. UniDiffuser provides two pretrained models, UniDiffuser-v0 and UniDiffuser-v1, trained on large-scale image-text datasets.

eSpark Learning is an AI-powered education platform designed to differentiate math, reading, and writing instruction for students. It extends core curriculum with adaptive, joyful practice aligned to daily teaching. The platform uses AI to personalize learning paths, ensuring practice is appropriately challenging and builds mastery step by step. Students explore subjects through topics they care about, fostering engagement and meaningful learning. eSpark provides immediate guidance and feedback, helping students strengthen understanding and build confidence. It integrates with existing assessments and offers explicit instruction, practice, and assessment, making it a comprehensive tool for teachers to manage diverse learning needs in their classrooms.

Felo is a free, multilingual AI search engine designed to provide real-time answers to user queries across various topics. Beyond just searching, Felo offers creative AI capabilities, allowing users to generate AI presentations, AI mind maps, and posters directly from their search results or input. This positions Felo as a versatile AI-powered answer engine, catering to both information retrieval and content creation needs. The platform emphasizes its ability to offer real-time responses and supports multiple languages, making it accessible to a global audience. Felo aims to streamline the process of finding information and transforming it into actionable or presentable formats.

DataRoot Labs is an AI R&D Center specializing in data science consulting and the development of AI-powered products. They offer 'R&D as a service' for startups, providing custom AI and ML solutions across various industries like healthcare, HR tech, retail, and logistics. Key offerings include free AI consulting sessions, detailed AI solution roadmaps, and fast MVP delivery within 8-12 weeks. They emphasize flexible pricing, full IP transfer, confidentiality, and even fundraising support for their clients. Their expertise spans technologies like Python, FastAPI, LangChain, Azure OpenAI, AWS Bedrock, and various vector databases, as demonstrated in their case studies.

GOAT.AI is a research lab dedicated to maximizing the strength of small models through Reinforcement Learning from Human Feedback (RLHF). The lab focuses on training compact yet robust models for domain-specific tasks, demonstrating that smaller models can outperform larger ones with appropriate datasets and reinforcement learning. Their research addresses critical problems in autonomous agents, including being stuck in loops, hallucinations (using Proximal Policy Optimization), and applying MetaRL techniques for experience. GOAT.AI is actively developing AI systems capable of understanding and generating natural language, while also tackling the software engineering challenges associated with large-scale AI models and millions of users.

WeKnora is an open-source, LLM-powered knowledge framework designed for enterprise-grade document understanding, semantic retrieval, and autonomous reasoning. It offers three core capabilities: RAG-based Quick Q&A for efficient lookups, a ReAct Agent that orchestrates retrieval, tools, and web search for complex multi-step tasks, and a Wiki Mode that distills documents into a self-maintaining, interlinked markdown knowledge base with an interactive knowledge graph. The platform supports auto-syncing knowledge from various sources like Feishu, Notion, and Yuque, handles over 10 document formats, and integrates with major LLM providers including OpenAI, DeepSeek, and Gemini. Its modular architecture allows for swapping LLMs, vector databases, and storage backends, ensuring complete data sovereignty with local and private cloud deployment options. WeKnora also provides comprehensive observability through Langfuse for agent reasoning and token usage.

Dessix is an AI-powered visual workspace designed to enhance the creative process for deep thinkers. It focuses on fostering shared human-AI alignment to effectively capture, organize, and expand ideas. The platform serves as an intelligent thinking partner, assisting users in developing and structuring their thoughts with the aid of artificial intelligence. Dessix aims to streamline the ideation process, making it easier for users to manage complex information and transform raw concepts into well-defined projects or insights. Its visual interface is tailored to support intuitive interaction and comprehensive idea management.

Mineflow is an AI-driven platform designed for mineral exploration, transforming raw exploration site data into highly accurate predictions of mineral deposit shapes and locations. It automates the extraction and organization of data from various file types like PDFs and Excel, eliminating manual transcription. The platform provides 3D visualization of drilling and non-drill datasets, auto-generated cross-sections, and the ability to export to standard geological formats. Mineflow supports off-the-shelf estimators like Kriging and Inverse Distance Weighting, and offers advanced features such as regulatory intelligence, searchable public datasets, and tenement monitoring for comprehensive exploration planning.

Question AI is a leading AI homework helper designed to assist students with a wide range of academic tasks, covering subjects from math and chemistry to literature and history. The tool utilizes state-of-the-art machine learning and advanced algorithms to deliver solutions with a 98% accuracy rate. Users can input questions by typing, uploading documents, or even images, and receive detailed, step-by-step answers. It offers 24/7 accessibility and multilingual support for over 100 languages, making it a versatile and convenient resource for students at all grade levels. Question AI aims to enhance learning by not only providing correct answers but also explaining the material in detail to help students understand and improve their grades.

Future Horizons Education specializes in driving AI innovation and strategy through comprehensive learning solutions and consultancy services. The platform offers courses designed to empower businesses and individuals, covering topics such as AI agents, strategy, and automations using tools like Zapier and n8n. Beyond digital learning, Future Horizons Education provides strategic governance and ethics training, and consultancy services for AI integration and workflow automation. Their goal is to help organizations navigate modern business transformation by bridging the gap between emerging technology and practical leadership, ensuring measurable growth and operational success through high-impact learning.

WAiNuT positions itself as the "one-stop AI shop" in the Netherlands, offering comprehensive services across AI consulting, training, and staffing. For businesses, they provide end-to-end AI consulting and implementation, including strategy development, Copilot rollouts, workflow automation, and AI Act compliance. Their training programs, using a proprietary CRAFTING method, aim to enhance AI literacy across organizations and offer role-based workshops for various departments. Additionally, WAiNuT specializes in AI staffing, recruiting and seconding AI and IT specialists, including AI-native trainees. They emphasize a practitioner-led approach, building and implementing solutions that go into production, and are vendor-independent, choosing the best AI platform for each client's needs.

Avathon, formerly MAANA, provides an AI platform designed to future-proof businesses by integrating intelligence and autonomy into physical operations. It equips industrial, logistics, and government partners with deep domain knowledge, powered by AI and agentic applications, to optimally manage assets, fleets, people, and networks. The platform is built on a computational knowledge graph, pre-populated with expertise across industrial and supply chain operations, and infused with advanced AI capabilities like machine vision, normal behavior modeling, generative AI, and natural language processing. This enables organizations to anticipate, diagnose, and act on operational challenges, leading to growth, cost savings, and resilience across the industrial economy.

SkillsTeq is an advanced e-learning and training platform that leverages AI to revolutionize corporate and sales training. It enables users to create interactive courses, sales and customer experience training, assessments, and feedback agents through a unique chat-based learning environment. The platform transforms company data into engaging e-learning modules, offering personalized learning experiences. Key features include AI-powered course generation, automated quiz and assessment creation, and immersive roleplay scenarios tailored to specific industry data. SkillsTeq aims to boost efficiency, improve knowledge retention, and prepare workforces for real-world scenarios by providing dynamic, adaptive learning journeys.

TOPIQ is an AI-powered news intelligence platform designed to keep users informed on any topic. It reads over 4 million articles daily in 55+ languages, providing personalized news signals and summaries directly to your inbox. Users can track companies, crypto, politics, sports, or any other topic, and TOPIQ's AI uses natural language processing, sentiment analysis, and multilingual capabilities to simplify complex information into actionable insights. This saves hours of research by delivering curated daily reports, making it ideal for professionals and enthusiasts who need to stay ahead of developments.