Today's Image & Video Generation: Fastest-Growing Projects — April 17, 2026
Today's the Image & Video Generation space, we've seen a surge in interest around multimodal foundation models and text-to-image generation tools. The top-growing repositories showcase innovative solutions for image understanding, editing, and generation, with many leveraging diffusion transformers and agent-based architectures. As AI capabilities continue to advance, these tools are likely to play a significant role in shaping the future of visual content creation.
JoyAI-Image, with a growth score of 70.41 and 1,833 stars, is a unified multimodal foundation model for image understanding, text-to-image generation, and instruction-guided image editing. Its rapid growth can be attributed to its comprehensive capabilities and potential applications in various industries.
ERNIE-Image, boasting a growth score of 61.83 and 221 stars, is an open text-to-image generation model developed by Baidu's ERNIE-Image team. With only 8B parameters, it achieves state-of-the-art performance among open-weight text-to-image models, making it an attractive choice for researchers and developers.
Ashim, with a growth score of 36.79 and 741 stars, is a Docker container housing over 30 tools and local AI capabilities for image processing tasks such as resizing, compressing, and upscaling. Its popularity stems from its self-contained nature, ensuring that users' images never leave their machines.
MOSS-VL, with a growth score of 25.11 and 218 stars, is the core multimodal model series within the OpenMOSS ecosystem, dedicated to visual understanding. As interest in multimodal AI grows, MOSS-VL's relevance and utility are driving its increasing adoption.
AlayaRenderer, sporting a growth score of 20.87 and 569 stars, is an AI-native renderer for games and virtual worlds developed by ShandaAI. Its unique approach to rendering has captured the attention of game developers and researchers exploring new ways to create immersive experiences.
Forge-film, with a growth score of 14.52 and 442 stars, is a multi-model DAG-driven parallel AI film generation engine. By allowing for simultaneous scene generation and optimizing execution graphs, Forge-film offers a novel solution for content creators seeking to accelerate their workflows.
Awesome-generative-models, featuring a growth score of 9.07 and 99 stars, provides weekly real-time updates on the latest diffusion generation models across various fields. This repository's utility as a knowledge hub has earned it a loyal following among researchers and enthusiasts tracking advancements in generative AI.
Pilipili-AutoVideo, with a growth score of 9.04 and 158 stars, is a fully automated AI video agent capable of generating videos with subtitles from text inputs. Its user-friendly nature and local deployment options have contributed to its growing popularity.
OpenMontage, boasting a growth score of 3.92 and an impressive 2,318 stars, is the world's first open-source agentic video production system. By transforming AI coding assistants into full-fledged video production studios, OpenMontage has democratized access to high-end video creation tools.
Lastly, GLD, with a growth score of 3.90 and 181 stars, is an official implementation of "Repurposing Geometric Foundation Models for Multi-view Diffusion". Although its growth rate is slower compared to other repositories on this list, GLD's contributions to the field of geometric foundation models have earned it recognition within the research community.
By analyzing these top-growing repositories in the Image & Video Generation space, we can identify key trends and areas of innovation that will likely shape the future of visual content creation.
JoyAI-Image, with a growth score of 70.41 and 1,833 stars, is a unified multimodal foundation model for image understanding, text-to-image generation, and instruction-guided image editing. Its rapid growth can be attributed to its comprehensive capabilities and potential applications in various industries.
ERNIE-Image, boasting a growth score of 61.83 and 221 stars, is an open text-to-image generation model developed by Baidu's ERNIE-Image team. With only 8B parameters, it achieves state-of-the-art performance among open-weight text-to-image models, making it an attractive choice for researchers and developers.
Ashim, with a growth score of 36.79 and 741 stars, is a Docker container housing over 30 tools and local AI capabilities for image processing tasks such as resizing, compressing, and upscaling. Its popularity stems from its self-contained nature, ensuring that users' images never leave their machines.
MOSS-VL, with a growth score of 25.11 and 218 stars, is the core multimodal model series within the OpenMOSS ecosystem, dedicated to visual understanding. As interest in multimodal AI grows, MOSS-VL's relevance and utility are driving its increasing adoption.
AlayaRenderer, sporting a growth score of 20.87 and 569 stars, is an AI-native renderer for games and virtual worlds developed by ShandaAI. Its unique approach to rendering has captured the attention of game developers and researchers exploring new ways to create immersive experiences.
Forge-film, with a growth score of 14.52 and 442 stars, is a multi-model DAG-driven parallel AI film generation engine. By allowing for simultaneous scene generation and optimizing execution graphs, Forge-film offers a novel solution for content creators seeking to accelerate their workflows.
Awesome-generative-models, featuring a growth score of 9.07 and 99 stars, provides weekly real-time updates on the latest diffusion generation models across various fields. This repository's utility as a knowledge hub has earned it a loyal following among researchers and enthusiasts tracking advancements in generative AI.
Pilipili-AutoVideo, with a growth score of 9.04 and 158 stars, is a fully automated AI video agent capable of generating videos with subtitles from text inputs. Its user-friendly nature and local deployment options have contributed to its growing popularity.
OpenMontage, boasting a growth score of 3.92 and an impressive 2,318 stars, is the world's first open-source agentic video production system. By transforming AI coding assistants into full-fledged video production studios, OpenMontage has democratized access to high-end video creation tools.
Lastly, GLD, with a growth score of 3.90 and 181 stars, is an official implementation of "Repurposing Geometric Foundation Models for Multi-view Diffusion". Although its growth rate is slower compared to other repositories on this list, GLD's contributions to the field of geometric foundation models have earned it recognition within the research community.
By analyzing these top-growing repositories in the Image & Video Generation space, we can identify key trends and areas of innovation that will likely shape the future of visual content creation.