Today's Image & Video Generation: Fastest-Growing Projects — June 21, 2026
This week, the Image & Video Generation space on GitHub continues to be dynamic, with several projects showing significant growth and engagement from developers and users alike. Among them, AI-driven video generation and image creation tools are particularly prominent, catering to diverse needs such as offline processing, web-based UIs, and specialized content creation.
lcy362/agnes-video-generator is a completely free AI video generation tool powered by Agnes AI, allowing users to input text ideas and automatically generate multi-scene videos through a Web UI. With a Growth Score of 38.78 and 62 stars, the project's steady increase in popularity can be attributed to its user-friendly interface and ability to create complex video content with minimal effort.
kadevin/ilab-gpt-conjure offers an AI image generation workbench for GPT-image-2 that supports Codex Responses and OpenAI-compatible API integration. The tool includes features like a shared gallery, multi-type quick chips, prompt templates, concurrent tasks, and local queue management. Its impressive Growth Score of 38.50 alongside 539 stars reflects its comprehensive feature set and strong community support.
techjarves/Portable-Diffusion is designed as a portable AI image generator that runs Stable Diffusion models locally on Windows, Linux, and macOS, with zero manual setup required. The tool automatically detects GPU/NPU acceleration for CUDA, ROCm, Vulkan, Metal, and OpenVINO environments. With 70 stars and a Growth Score of 24.10, its ease-of-use and multi-platform support have likely contributed to its growing popularity.
helloianneo/ian-xiaohei-illustrations provides a skill for generating Chinese-style illustrations, featuring hand-drawn images with red-orange-blue annotations. The repository has gained considerable attention, with 5,433 stars and a Growth Score of 20.97, highlighting its appeal to content creators seeking unique visual styles.
techjarves/Local-AI-Image-Generator is an offline AI image generation studio for Windows that supports Stable Diffusion models locally without manual setup, auto-configuring CUDA for Nvidia GPUs and Vulkan for AMD/Intel Arc cards. The tool's 222 stars and a Growth Score of 19.32 indicate its growing traction among users looking to generate images without internet connectivity.
nv-tlabs/dvlt is the official implementation of Déjà View, which uses Looping Transformers for multi-view 3D reconstruction. Despite having fewer commits in the last month (only 3), it still maintains a strong Growth Score of 8.98 and 335 stars due to its technical novelty and relevance in advanced image processing.
helloianneo/ian-xiaohei-scenes introduces Xiaohei 2.0 Codex Skill for generating Chinese real-object article illustrations and long-scroll story images, catering specifically to a niche audience interested in detailed storytelling through visuals. Its Growth Score of 8.97 and 236 stars suggest it has found a dedicated user base appreciating its unique content creation capabilities.
Lekien082/AI-Photo-Background-Remover is an easy-to-use desktop utility powered by AI to automatically remove backgrounds from photos locally, offline, and in high resolution. With a Growth Score of 7.75 and 75 stars, the tool's practicality and ease-of-use are likely driving its growth among users seeking quick photo editing solutions.
Guo-chunyu/Chaning.G-s-Lrlab integrates large language models with neural network-level color grading algorithms to bridge camera RAW folders/photo collections directly with AI inspiration for Lightroom presets generation. The project's 100 stars and a Growth Score of 3.52 reflect its specialized appeal among professional photographers looking to streamline their post-processing workflow.
Lastly, MSALab-PKU/LoomVideo is the official implementation focusing on unifying multimodal inputs into video generation and editing processes. With 62 stars and a Growth Score of 3.05, it suggests ongoing interest from researchers and developers in advanced multimedia integration capabilities.
Overall, these tools highlight the diverse applications and growing demand for AI-driven image and video creation across various user segments, from casual content creators to professional photographers and data scientists.
lcy362/agnes-video-generator is a completely free AI video generation tool powered by Agnes AI, allowing users to input text ideas and automatically generate multi-scene videos through a Web UI. With a Growth Score of 38.78 and 62 stars, the project's steady increase in popularity can be attributed to its user-friendly interface and ability to create complex video content with minimal effort.
kadevin/ilab-gpt-conjure offers an AI image generation workbench for GPT-image-2 that supports Codex Responses and OpenAI-compatible API integration. The tool includes features like a shared gallery, multi-type quick chips, prompt templates, concurrent tasks, and local queue management. Its impressive Growth Score of 38.50 alongside 539 stars reflects its comprehensive feature set and strong community support.
techjarves/Portable-Diffusion is designed as a portable AI image generator that runs Stable Diffusion models locally on Windows, Linux, and macOS, with zero manual setup required. The tool automatically detects GPU/NPU acceleration for CUDA, ROCm, Vulkan, Metal, and OpenVINO environments. With 70 stars and a Growth Score of 24.10, its ease-of-use and multi-platform support have likely contributed to its growing popularity.
helloianneo/ian-xiaohei-illustrations provides a skill for generating Chinese-style illustrations, featuring hand-drawn images with red-orange-blue annotations. The repository has gained considerable attention, with 5,433 stars and a Growth Score of 20.97, highlighting its appeal to content creators seeking unique visual styles.
techjarves/Local-AI-Image-Generator is an offline AI image generation studio for Windows that supports Stable Diffusion models locally without manual setup, auto-configuring CUDA for Nvidia GPUs and Vulkan for AMD/Intel Arc cards. The tool's 222 stars and a Growth Score of 19.32 indicate its growing traction among users looking to generate images without internet connectivity.
nv-tlabs/dvlt is the official implementation of Déjà View, which uses Looping Transformers for multi-view 3D reconstruction. Despite having fewer commits in the last month (only 3), it still maintains a strong Growth Score of 8.98 and 335 stars due to its technical novelty and relevance in advanced image processing.
helloianneo/ian-xiaohei-scenes introduces Xiaohei 2.0 Codex Skill for generating Chinese real-object article illustrations and long-scroll story images, catering specifically to a niche audience interested in detailed storytelling through visuals. Its Growth Score of 8.97 and 236 stars suggest it has found a dedicated user base appreciating its unique content creation capabilities.
Lekien082/AI-Photo-Background-Remover is an easy-to-use desktop utility powered by AI to automatically remove backgrounds from photos locally, offline, and in high resolution. With a Growth Score of 7.75 and 75 stars, the tool's practicality and ease-of-use are likely driving its growth among users seeking quick photo editing solutions.
Guo-chunyu/Chaning.G-s-Lrlab integrates large language models with neural network-level color grading algorithms to bridge camera RAW folders/photo collections directly with AI inspiration for Lightroom presets generation. The project's 100 stars and a Growth Score of 3.52 reflect its specialized appeal among professional photographers looking to streamline their post-processing workflow.
Lastly, MSALab-PKU/LoomVideo is the official implementation focusing on unifying multimodal inputs into video generation and editing processes. With 62 stars and a Growth Score of 3.05, it suggests ongoing interest from researchers and developers in advanced multimedia integration capabilities.
Overall, these tools highlight the diverse applications and growing demand for AI-driven image and video creation across various user segments, from casual content creators to professional photographers and data scientists.