Today's Image & Video Generation: Fastest-Growing Projects — May 23, 2026
Today's the Image & Video Generation space, we see continued growth and innovation across various tools that cater to different use cases, from professional image editing to creative content generation. One standout repository is bytedance/Lance, a lightweight unified multimodal model for comprehensive tasks related to image and video understanding, generation, and editing.
bytedance/Lance, with a Growth Score of 66.60, stands out as one of the most active projects in the category despite having no star ratings yet. Its focus on providing an all-in-one solution for multimodal tasks suggests it is gaining traction among developers looking to integrate advanced image and video capabilities into their applications.
BasZ4ll/Stable-Diffusion-WebUI offers a comprehensive suite of tools for Stable Diffusion, including setup guides and optimizations for various configurations like low VRAM optimization and xformers command line arguments. With 399 stars and a Growth Score of 35.25, this repository is growing steadily due to its detailed documentation and community support for optimizing performance on different hardware setups.
mrslimslim/gpt-image-canvas provides a local professional AI canvas built with tldraw, enabling users to create complex images using AI-driven tools within the browser. The project has seen significant development activity with 83 commits in the last month, contributing to its Growth Score of 24.72 and accumulating 513 stars as users appreciate the ease of use and robust feature set for professional image creation.
helloianneo/ian-handdrawn-ppt is a repository that focuses on generating high-quality hand-drawn presentation slides and cover images in PNG format, catering specifically to Chinese technical presentations. With 873 stars and a Growth Score of 20.20, this tool's popularity likely stems from its specialized use case and the detailed documentation provided for users looking to enhance their visual communication skills.
RareSense/Nova3D offers an editable, part-aware 3D generation system that can create models based on text descriptions or reference images, with a client available at nova3d.xyz. The project's active development (26 commits in the last month) and its Growth Score of 13.82 alongside 227 stars indicate growing interest from developers interested in integrating advanced 3D generation capabilities into their projects.
ailuntx/Thinking-with-Visual-Primitives is an archived snapshot of a project that explores visual primitives for AI image creation, though it has received limited attention with only 109 stars and a Growth Score of 5.96. The lack of recent activity suggests the repository may no longer be under active development.
Hao0321/ai-media-generator is a skill package designed to assist in crafting high-quality AI-generated media prompts across multiple platforms, including images, videos, and music. With 28 stars and a Growth Score of 5.90, this tool's growth reflects its utility for users who want to streamline the process of generating complex media content through automation.
gpt-img-2/ai-image-prompt-cookbook is another repository that provides Chinese-language AI image prompt libraries covering various use cases such as e-commerce main images and social media posts. The project has garnered 26 stars with a Growth Score of 4.06, indicating moderate interest from developers looking to leverage pre-defined prompts for generating specific types of visual content.
jieyefriic/nbcraft is a multi-backend image and video generation command-line interface (CLI) that supports several AI platforms like Gemini, DashScope, Volcengine Ark, and OpenAI gpt-image-2. With 153 stars and a Growth Score of 3.78, the project's steady growth can be attributed to its versatility in supporting multiple backends for image and video generation tasks.
jiangmuran/claude-image is a skill package designed to teach Claude Code/Codex how to effectively use GPT Image 2 by providing intent-first prompting, surgical edits, parallel batching, and visual self-verification features. The project has accumulated 59 stars with a Growth Score of 1.98, reflecting its specialized focus on enhancing the capabilities of AI agents in generating high-quality images through advanced prompting techniques.
Today's radar highlights a diverse range of tools addressing various aspects of image and video generation, from lightweight multimodal models to specialized presentation generators and multi-backend CLI interfaces, each contributing to the vibrant ecosystem of AI-driven visual content creation.
bytedance/Lance, with a Growth Score of 66.60, stands out as one of the most active projects in the category despite having no star ratings yet. Its focus on providing an all-in-one solution for multimodal tasks suggests it is gaining traction among developers looking to integrate advanced image and video capabilities into their applications.
BasZ4ll/Stable-Diffusion-WebUI offers a comprehensive suite of tools for Stable Diffusion, including setup guides and optimizations for various configurations like low VRAM optimization and xformers command line arguments. With 399 stars and a Growth Score of 35.25, this repository is growing steadily due to its detailed documentation and community support for optimizing performance on different hardware setups.
mrslimslim/gpt-image-canvas provides a local professional AI canvas built with tldraw, enabling users to create complex images using AI-driven tools within the browser. The project has seen significant development activity with 83 commits in the last month, contributing to its Growth Score of 24.72 and accumulating 513 stars as users appreciate the ease of use and robust feature set for professional image creation.
helloianneo/ian-handdrawn-ppt is a repository that focuses on generating high-quality hand-drawn presentation slides and cover images in PNG format, catering specifically to Chinese technical presentations. With 873 stars and a Growth Score of 20.20, this tool's popularity likely stems from its specialized use case and the detailed documentation provided for users looking to enhance their visual communication skills.
RareSense/Nova3D offers an editable, part-aware 3D generation system that can create models based on text descriptions or reference images, with a client available at nova3d.xyz. The project's active development (26 commits in the last month) and its Growth Score of 13.82 alongside 227 stars indicate growing interest from developers interested in integrating advanced 3D generation capabilities into their projects.
ailuntx/Thinking-with-Visual-Primitives is an archived snapshot of a project that explores visual primitives for AI image creation, though it has received limited attention with only 109 stars and a Growth Score of 5.96. The lack of recent activity suggests the repository may no longer be under active development.
Hao0321/ai-media-generator is a skill package designed to assist in crafting high-quality AI-generated media prompts across multiple platforms, including images, videos, and music. With 28 stars and a Growth Score of 5.90, this tool's growth reflects its utility for users who want to streamline the process of generating complex media content through automation.
gpt-img-2/ai-image-prompt-cookbook is another repository that provides Chinese-language AI image prompt libraries covering various use cases such as e-commerce main images and social media posts. The project has garnered 26 stars with a Growth Score of 4.06, indicating moderate interest from developers looking to leverage pre-defined prompts for generating specific types of visual content.
jieyefriic/nbcraft is a multi-backend image and video generation command-line interface (CLI) that supports several AI platforms like Gemini, DashScope, Volcengine Ark, and OpenAI gpt-image-2. With 153 stars and a Growth Score of 3.78, the project's steady growth can be attributed to its versatility in supporting multiple backends for image and video generation tasks.
jiangmuran/claude-image is a skill package designed to teach Claude Code/Codex how to effectively use GPT Image 2 by providing intent-first prompting, surgical edits, parallel batching, and visual self-verification features. The project has accumulated 59 stars with a Growth Score of 1.98, reflecting its specialized focus on enhancing the capabilities of AI agents in generating high-quality images through advanced prompting techniques.
Today's radar highlights a diverse range of tools addressing various aspects of image and video generation, from lightweight multimodal models to specialized presentation generators and multi-backend CLI interfaces, each contributing to the vibrant ecosystem of AI-driven visual content creation.