PullRepo

Daily radar for the fastest-growing AI tools & repos

Today's Image & Video Generation: Fastest-Growing Projects — May 19, 2026

Today's the Image & Video Generation space, there's a noticeable trend towards more accessible and versatile AI-driven solutions that cater to both professional and casual users. Among these offerings, several repositories have seen significant growth, driven by their innovative approaches and community engagement.

CookSleep/gpt_image_playground is an interface-based tool built around OpenAI’s gpt-image-2 API, enabling users to generate and edit images with ease. Its high Growth Score of 61.13 and a substantial number of stars (1,406) indicate its popularity among developers seeking powerful yet straightforward image generation capabilities.

bytedance/Lance is described as a lightweight native unified multimodal model for understanding, generating, and editing both images and videos. Despite having fewer stars compared to some other entries, it has seen steady growth with 32.75 Growth Score and frequent commits over the last month, suggesting active development and interest from its user base.

mrslimslim/gpt-image-canvas offers a local AI-driven canvas that leverages tldraw for professional image generation and editing tasks. With a notable Growth Score of 23.90 and 371 stars, this tool is growing due to its unique approach to integrating AI within an intuitive drawing interface.

helloianneo/ian-handdrawn-ppt focuses on generating Chinese-style hand-drawn PPT pages, including cover slides and body images for presentations in PNG format. Its high star count (850) reflects the demand for such specialized content creation tools among users looking for distinctive visual elements.

lide-jun/ima2-gen is a minimalist command-line interface (CLI) along with a web UI that leverages OpenAI’s GPT Image 2 API to generate images from text prompts or other images. The tool supports both paid and free authentication methods, which likely contributes to its growth score of 17.11 alongside an active development cycle.

RareSense/Nova3D is an open-source client for nova3d.xyz, enabling part-aware 3D generation based on text descriptions or reference images. Although it lacks star ratings, the project’s consistent coding activity and innovative approach make it a notable entry in Today's radar.

BasZ4ll/Stable-Diffusion-WebUI provides extensive resources and setup guides for users looking to implement Stable Diffusion WebUI on their local machines. Its comprehensive documentation and community-driven updates have garnered 435 stars, reflecting its importance among enthusiasts and developers working with image generation technologies.

AITOOLS101/ai-image-prompt-cookbook is a repository dedicated to Chinese AI image prompt phrases covering various categories like fashion, children's wear, and commercial photography. With a Growth Score of 7.88 and modest star count (24), its relevance in the specialized content creation market helps explain its steady growth.

kevinchin12/ImageLens offers an innovative Chrome extension that allows users to reverse-engineer prompts from web images and test their effectiveness instantly. This unique approach, combined with frequent commits over the past month, contributes to its growing popularity among those interested in understanding AI-generated image prompts better.

Finally, Visko-Platform/VEFX-Bench presents a comprehensive benchmarking tool for evaluating generic video editing and visual effects systems. Its 4.91 Growth Score alongside 204 stars indicates ongoing interest from researchers and developers looking for standardized evaluation methods in the video generation space.
Back to all reports