Today's Image & Video Generation: Fastest-Growing Projects — August 23, 2026
Today's the Image & Video Generation space on GitHub, there's a notable surge in projects that enhance visual creativity and editing capabilities without relying heavily on external APIs like OpenAI’s. These tools are gaining traction for their innovative approaches to image generation and manipulation, offering users more flexibility and control over their creative processes.
The project "dsh-image-gen" by shanliuling stands out with a growth score of 21.33 and 133 stars. This tool allows users to generate images directly within DeepSeek Harness chats, making it an accessible solution for those who want to integrate image creation into their communication workflows seamlessly.
"six-nut/PocketMen-with-you" has garnered attention with a growth score of 18.00 and 90 stars. The project enables the creation of high-fidelity companions using local neural editing techniques, requiring only two or more photos and no OpenAI API key. This makes it an attractive option for developers looking to incorporate advanced image editing functionalities into their applications without external dependencies.
"4DAnyone" by ant-research has a growth score of 15.85 and over 300 stars. The tool is designed to create realistic 4D representations from casual monocular video inputs, showcasing its potential in transforming how users interact with visual content through dynamic and lifelike recreations.
"dacnay816y62-hub/photo-revival" has seen a growth score of 14.86 and an impressive 474 stars. This Codex skill transforms everyday photos into poetic, hand-drawn illustrations, appealing to artists and designers who seek unique visual styles for their projects.
The repository "muyang-illustration-skills" by yokel1121 has a growth score of 12.29 and 190 stars. It provides users with 25 fixed illustration styles, prompt recipes, and visual references, making it easier to create consistent and high-quality illustrations within Codex.
"Sateezg/codex-bridge" boasts a growth score of 10.89 and 251 stars. This tool offers image generation and GPT-5 subagents through the Codex CLI without requiring an OpenAI API key, making it accessible for developers to integrate advanced AI capabilities into their projects.
The project "codex-image-context-runtime" by shixinnt has a growth score of 8.67 and 100 stars. It includes a plugin and local MCP runtime that enables context-bounded image generation and inspection, offering users greater control over the creation process within Codex environments.
"oil-oil/dsh-vision," with a growth score of 7.67 and 84 stars, enhances near-native image understanding for DeepSeek Harness, making it easier to analyze and work with images directly in chat interfaces.
The tool "xxg-portrait-rebuild-light" by moskoo has a growth score of 6.67 and 73 stars. It focuses on reconstructing lighting elements and restoring skin texture in portrait photos using a director-led workflow, providing photographers and artists with advanced editing tools for portraits.
Finally, the project "HelixWorld" by NoizAI has a growth score of 6.62 and 199 stars. This real-time interactive audio-visual world model allows users to create immersive experiences that combine visual and auditory elements in a dynamic setting, appealing to developers interested in creating engaging multimedia applications.
These tools collectively demonstrate the diversity and innovation within the image and video generation space on GitHub this week, catering to various needs from artistic creativity to technical development.
The project "dsh-image-gen" by shanliuling stands out with a growth score of 21.33 and 133 stars. This tool allows users to generate images directly within DeepSeek Harness chats, making it an accessible solution for those who want to integrate image creation into their communication workflows seamlessly.
"six-nut/PocketMen-with-you" has garnered attention with a growth score of 18.00 and 90 stars. The project enables the creation of high-fidelity companions using local neural editing techniques, requiring only two or more photos and no OpenAI API key. This makes it an attractive option for developers looking to incorporate advanced image editing functionalities into their applications without external dependencies.
"4DAnyone" by ant-research has a growth score of 15.85 and over 300 stars. The tool is designed to create realistic 4D representations from casual monocular video inputs, showcasing its potential in transforming how users interact with visual content through dynamic and lifelike recreations.
"dacnay816y62-hub/photo-revival" has seen a growth score of 14.86 and an impressive 474 stars. This Codex skill transforms everyday photos into poetic, hand-drawn illustrations, appealing to artists and designers who seek unique visual styles for their projects.
The repository "muyang-illustration-skills" by yokel1121 has a growth score of 12.29 and 190 stars. It provides users with 25 fixed illustration styles, prompt recipes, and visual references, making it easier to create consistent and high-quality illustrations within Codex.
"Sateezg/codex-bridge" boasts a growth score of 10.89 and 251 stars. This tool offers image generation and GPT-5 subagents through the Codex CLI without requiring an OpenAI API key, making it accessible for developers to integrate advanced AI capabilities into their projects.
The project "codex-image-context-runtime" by shixinnt has a growth score of 8.67 and 100 stars. It includes a plugin and local MCP runtime that enables context-bounded image generation and inspection, offering users greater control over the creation process within Codex environments.
"oil-oil/dsh-vision," with a growth score of 7.67 and 84 stars, enhances near-native image understanding for DeepSeek Harness, making it easier to analyze and work with images directly in chat interfaces.
The tool "xxg-portrait-rebuild-light" by moskoo has a growth score of 6.67 and 73 stars. It focuses on reconstructing lighting elements and restoring skin texture in portrait photos using a director-led workflow, providing photographers and artists with advanced editing tools for portraits.
Finally, the project "HelixWorld" by NoizAI has a growth score of 6.62 and 199 stars. This real-time interactive audio-visual world model allows users to create immersive experiences that combine visual and auditory elements in a dynamic setting, appealing to developers interested in creating engaging multimedia applications.
These tools collectively demonstrate the diversity and innovation within the image and video generation space on GitHub this week, catering to various needs from artistic creativity to technical development.