Generative AI

Index-AniSora is a powerful open-source framework designed specifically for high-quality anime video generation and animation production. The system features a comprehensive data processing pipeline, a controllable generation model with spatiotemporal masking, and a specialized evaluation benchmark. It supports diverse creative tasks including character 3D video generation, style transfer, and multimodal guidance for precise motion control.

ArcReel is an open-source video generation workbench powered by AI Agents, designed to automate the production process from novel scripts to short video clips. The platform supports multi-vendor image and video generation models, ensuring character consistency and visual coherence through a multi-agent architecture. Users can manage projects, track generation costs, and export Jianying drafts via a visual interface for an efficient video creation experience.

VoxCPM2 is a tokenizer-free, 2B parameter text-to-speech system that utilizes a diffusion autoregressive architecture to generate high-quality, expressive audio. The model supports 30 languages and offers advanced capabilities including natural-language voice design and controllable voice cloning. It is fully open-source under the Apache-2.0 license and provides production-ready features like real-time streaming and high-fidelity 48kHz output.

ERNIE-Image is an open-source text-to-image model developed by Baidu based on a single-stream Diffusion Transformer architecture. The model is equipped with a lightweight prompt enhancer capable of transforming short inputs into structurally rich, detailed descriptions. With an 8B parameter scale, it demonstrates industry-leading performance in text rendering and instruction following while maintaining efficient deployment.

OpenMontage is an open-source, agentic system that transforms AI coding assistants into comprehensive video production studios. It automates the entire creative workflow, including research, scripting, asset generation, editing, and final composition. The platform supports both AI-generated visuals and real-footage documentary montages using a variety of free and premium tools.

JoyAI-Image is a unified multimodal foundation model that integrates an 8B Multimodal Large Language Model with a 16B Multimodal Diffusion Transformer to support image understanding, generation, and editing. The model utilizes a closed-loop collaboration between understanding and generation to enhance spatial reasoning and controllable editing capabilities. It provides a scalable training pipeline and supports advanced features like multi-view generation and precise spatial manipulation.