PullRepo

Daily radar for the fastest-growing AI tools & repos

Today's Image & Video Generation: Fastest-Growing Projects — November 05, 2026

Today's the Image & Video Generation space, we see a strong trend towards local desktop applications and comprehensive pose libraries that cater to AI image generation tools like Midjourney and ComfyUI. These resources are designed to enhance creativity and streamline workflows for artists working with text-to-image and text-to-video technologies. One standout project is silverdugongenhance/PromptReel, a free desktop app that converts text prompts into short video clips on local Windows machines without the need for subscriptions or cloud credits.

silverdugongenhance/PromptReel has seen impressive growth this week with a score of 97.00 and currently boasts 104 stars. The tool's appeal lies in its ease-of-use, allowing users to generate video clips on their own PCs without the hassle of waiting for cloud-based resources.

wquguru/shisei is another notable project, offering a free CC0 mannequin pose reference library that serves as an invaluable resource for AI image tools such as Nano Banana and Midjourney. This repository includes over 250 poses across five different angles in line art format, with names provided in multiple languages including Japanese, Chinese, Korean, and English.

The project's growth score of 92.64 and 436 stars indicate its popularity among artists and developers looking for comprehensive pose libraries to enhance their AI-generated images. The availability of poses in various angles and language support makes it a versatile tool for different user bases.

SAOS-3624/dsh-video-storyboard offers an innovative solution with scene-aware storyboards for DeepSeek Harness, presenting 9-16 numbered cells that include timestamps, transcripts, and audio envelopes within a single image. This allows for efficient video planning and editing directly from the storyboard.

With a growth score of 56.00 and 430 stars, dsh-video-storyboard stands out due to its unique approach in integrating multiple elements into one cohesive image format. The tool's ability to enhance workflow efficiency by providing visual representations with comprehensive data points is likely driving its popularity among video creators.

Deeyu/ai-costume-gallery-zh provides a rich repository of AI costume reference images, including both realistic and animated clothing designs, facial shapes, pose line drawings, and depth maps. This resource supports multi-lingual browsing and image download capabilities, catering to the needs of AI artists and designers.

The project's growth score of 33.79 and 116 stars reflect its growing importance in providing a wide range of costume references for AI-driven creative projects. The comprehensive nature of the gallery, along with its multi-language support, likely contributes to its increasing star count as it aids users in generating diverse and detailed characters.

langchain-ai/design-agent is an open-source design agent that creates on-brand visuals such as stills, decks, diagrams, charts, and motion graphics based on chat inputs. It leverages Deep Agents technology for seamless integration into creative workflows.

With a growth score of 30.75 and 561 stars, langchain-ai/design-agent is gaining traction due to its ability to produce professional-grade design assets from conversational inputs. The tool's versatility in generating various types of visual content makes it an attractive solution for designers looking to streamline their creative processes.

Brilliant666/ai-costume-gallery-cn mirrors Deeyu’s project but focuses on a static gallery with no dependencies, allowing for easy access and use of costume references without the need for additional software. This repository supports Chinese language searches and multi-angle browsing capabilities.

The project's growth score of 26.00 and 139 stars suggest it is becoming increasingly popular among AI artists seeking to enhance their character designs with realistic and animated clothing options. The lack of dependencies and comprehensive costume references likely contribute to its growing user base.

paipiange/nvzhuang offers a substantial collection of 1407 AI-generated women's clothing reference images, complete with a virtual try-on website supporting both Chinese and English language searches. This project also integrates various image generation APIs such as OpenAI Images, Gemini, and custom Base URLs.

With an impressive growth score of 18.96 and 2,192 stars, paipiange/nvzhuang's extensive database and user-friendly interface are driving its popularity among artists and designers looking for detailed clothing references. The integration of multiple APIs and language support enhances the tool’s utility across different platforms.

ItsSNOWMOBILE/aftertake is an open-source AI video studio designed to run on personal GPUs, offering functionalities such as video generation, recasting, enhancement, and cutting using open-weight models and native ComfyUI nodes.

The project's growth score of 17.95 and 82 stars indicate its growing relevance in providing a local solution for advanced video editing tasks. The ability to leverage GPU resources for
Back to all reports