Aqaba.ai is a cloud GPU computing service designed to accelerate AI research and development by providing instant access to powerful GPUs such as H100s, A100s, and RTX cards. The platform enables developers to fine-tune the latest large language models, train custom AI models, and efficiently run AI workloads in a scalable environment. What sets Aqaba.ai apart is its commitment to sustainability, powering its infrastructure with renewable energy. Users get dedicated GPU instances exclusively allocated to them, allowing full control and maximizing performance without resource sharing or interruptions due to idling. With an easy-to-use prepaid credit system and support through live Discord and email, Aqaba.ai is trusted by over 1,000 AI developers worldwide.

TrainEngine.ai specializes in enabling users to train, fine-tune, and deploy AI models effortlessly. The platform is designed to support the development and application of image models, allowing for the generation of AI art, customization of models, and seamless integration into various workflows. With its intuitive interface and robust capabilities, TrainEngine.ai is an ideal choice for artists, data scientists, and AI enthusiasts looking to harness the power of machine learning for their creative projects.

Tromero is a cutting-edge AI training and hosting platform that leverages blockchain technology to provide enterprises with a competitive edge. It allows users to train and deploy machine learning models more efficiently and at reduced costs. Designed for scalability and ease of use, Tromero supports GPU clusters and offers various tools for performance evaluation, benchmarking, and real-time monitoring. Whether you're looking to train complex models or host AI applications, Tromero provides a comprehensive framework maximizing resource utilization and minimizing expenses.

This repository provides an end-to-end reinforcement learning framework for StarCraft II gameplay research. The core agent uses Proximal Policy Optimization (PPO) to learn policy networks that interpret observation data from the PySC2 environment and output precise in-game actions. Developers can configure neural network layers, reward shaping, and training schedules to optimize performance. The system supports multiprocessing for efficient sample collection, logging utilities for monitoring training curves, and evaluation scripts for running trained policies against scripted or built-in AI opponents. The codebase is written in Python and leverages TensorFlow for model definition and optimization. Users can extend components such as custom reward functions, state preprocessing, or network architectures to suit specific research objectives.