Today's Image & Video Generation: Fastest-Growing Projects — April 22, 2026
Today's Image & Video Generation, we're seeing a surge of interest in multimodal foundation models and text-to-image generation tools. The top-growing repositories are leveraging advancements in AI to push the boundaries of visual content creation, with a focus on unified models that can handle various tasks such as image understanding, editing, and generation.
jd-opensource/JoyAI-Image has taken the lead with a growth score of 58.43 and 1,941 stars. This repository offers JoyAI-Image, a unified multimodal foundation model for image understanding, text-to-image generation, and instruction-guided image editing. Its growing popularity can be attributed to its versatility and ability to perform multiple tasks under one umbrella.
baidu/ERNIE-Image follows closely with a growth score of 32.62 and 327 stars. ERNIE-Image is an open text-to-image generation model developed by the ERNIE-Image team at Baidu, built on a single-stream Diffusion Transformer (DiT) that achieves state-of-the-art performance among open-weight text-to-image models. Its growth can be attributed to its impressive performance and accessibility as an open-source tool.
ashim-hq/ashim boasts a growth score of 32.06 and 855 stars, offering a comprehensive Docker container with over 30 tools for image processing, including resize, compress, remove backgrounds, upscale, OCR, and more. Its growing popularity stems from its convenience and local AI capabilities that ensure users' images never leave their machine.
kangarooking/design-image-studio has gained significant attention with a growth score of 23.88 and 80 stars. This repository aims to turn vague visual demands into high-quality design graphics, providing a unique solution for designers and creators. Its growth can be attributed to its innovative approach to bridging the gap between creative ideas and tangible designs.
YouMind-OpenLab/awesome-gpt-image-2 has also seen significant growth with a score of 20.67 and 119 stars. This repository curates a library of prompts for OpenAI's GPT Image 2 model, featuring pixel-perfect text rendering, cross-image consistency, and commercial-grade illustration in 16 languages. Its popularity stems from its utility as a free and open-source resource for creatives.
Other notable repositories include OpenMOSS/MOSS-VL with a growth score of 17.32 and 227 stars, offering a core multimodal model series within the OpenMOSS ecosystem dedicated to visual understanding; ShandaAI/AlayaRenderer with a growth score of 15.68 and 570 stars, providing an AI-native renderer for games and virtual worlds; F-R-L/forge-film with a growth score of 13.47 and 526 stars, featuring a multi-model DAG-driven parallel AI film generation engine; gulucaptain/Camera-Transformer-1 with a growth score of 11.25 and 117 stars, enabling precise camera control for video generation; and jiangchaokang/awesome-generative-models with a growth score of 8.00 and 100 stars, offering weekly real-time updates on the latest diffusion generation models.
These repositories showcase the rapid advancements being made in Image & Video Generation, from unified multimodal foundation models to specialized tools for text-to-image generation, image processing, and film creation.
jd-opensource/JoyAI-Image has taken the lead with a growth score of 58.43 and 1,941 stars. This repository offers JoyAI-Image, a unified multimodal foundation model for image understanding, text-to-image generation, and instruction-guided image editing. Its growing popularity can be attributed to its versatility and ability to perform multiple tasks under one umbrella.
baidu/ERNIE-Image follows closely with a growth score of 32.62 and 327 stars. ERNIE-Image is an open text-to-image generation model developed by the ERNIE-Image team at Baidu, built on a single-stream Diffusion Transformer (DiT) that achieves state-of-the-art performance among open-weight text-to-image models. Its growth can be attributed to its impressive performance and accessibility as an open-source tool.
ashim-hq/ashim boasts a growth score of 32.06 and 855 stars, offering a comprehensive Docker container with over 30 tools for image processing, including resize, compress, remove backgrounds, upscale, OCR, and more. Its growing popularity stems from its convenience and local AI capabilities that ensure users' images never leave their machine.
kangarooking/design-image-studio has gained significant attention with a growth score of 23.88 and 80 stars. This repository aims to turn vague visual demands into high-quality design graphics, providing a unique solution for designers and creators. Its growth can be attributed to its innovative approach to bridging the gap between creative ideas and tangible designs.
YouMind-OpenLab/awesome-gpt-image-2 has also seen significant growth with a score of 20.67 and 119 stars. This repository curates a library of prompts for OpenAI's GPT Image 2 model, featuring pixel-perfect text rendering, cross-image consistency, and commercial-grade illustration in 16 languages. Its popularity stems from its utility as a free and open-source resource for creatives.
Other notable repositories include OpenMOSS/MOSS-VL with a growth score of 17.32 and 227 stars, offering a core multimodal model series within the OpenMOSS ecosystem dedicated to visual understanding; ShandaAI/AlayaRenderer with a growth score of 15.68 and 570 stars, providing an AI-native renderer for games and virtual worlds; F-R-L/forge-film with a growth score of 13.47 and 526 stars, featuring a multi-model DAG-driven parallel AI film generation engine; gulucaptain/Camera-Transformer-1 with a growth score of 11.25 and 117 stars, enabling precise camera control for video generation; and jiangchaokang/awesome-generative-models with a growth score of 8.00 and 100 stars, offering weekly real-time updates on the latest diffusion generation models.
These repositories showcase the rapid advancements being made in Image & Video Generation, from unified multimodal foundation models to specialized tools for text-to-image generation, image processing, and film creation.