<p>Animation production is increasingly shaped by collaborative workflows and compute-intensive pipelines, where content creation, model inference, and high-quality rendering are often conducted in parallel by geographically distributed users. However, conventional animation tools are largely built around standalone environments, making it difficult to scale computation resources or support efficient multi-user collaboration. Cloud platforms offer elastic resources and distributed execution capabilities, yet integrating animation generation, rendering, and collaboration into a unified cloud-based system remains challenging. This paper presents a cloud-based collaborative animation generation system that integrates open-source generative models with distributed rendering and resource scheduling services. The proposed framework enables multiple users to concurrently submit animation tasks, which are decomposed into generation and rendering subtasks and executed across cloud nodes through a coordinated orchestration layer. A distributed rendering pipeline is developed to support scalable execution of Blender-based workloads, while a resource scheduling mechanism is designed to manage heterogeneous GPU resources and dynamically allocate tasks according to workload conditions. Extensive experiments are conducted to evaluate both system-level and animation-oriented performance. The results demonstrate that the proposed system can effectively support concurrent animation creation while achieving stable performance gains and improved resource efficiency in cloud environments.</p>