Ketu
Kuaishou has launched Ketu Large Model, which provides Chinese text-to-image and AI image customization capabilities. It is built with a variety of stylized templates and a work square, supports one-click same-style creation, and is targeted at social avatar, creative illustration and personal image creation scenarios.
Updated: 2026-10-26
What is Ketu?
KOLORS, a large-model image product launched by Kuaishou, offers two main lines on its official website: AI creation and AI image customization. Compared with pure text-to-image tools, it places stylized templates and a work square in prominent positions, allowing users to preview the effects first and then generate the same style with one click, which significantly lowers the entry threshold. Its overall style is biased towards social communication, making it suitable for quickly creating finished products that can be directly shared, rather than staying on experimental sketches.
Who is it for?
- Social media users: Create avatars, backgrounds and personalized profile photos for daily sharing and interaction with friends, valuing fun and the speed of generating great results.
- Short video operators: Create image materials for video covers and account visuals, and need to maintain a stable update frequency and a unified visual tone.
- E-commerce and live streaming teams: For creative attempts and preliminary proposals of product images and scene images, quickly verifying whether the visual direction is feasible.
- Enthusiasts of design and content direction: Used for style exploration and personal practice, gradually building an understanding of prompts through replication and revision.
Key features
6Chinese Text-to-Image
The official version states that it can understand Chinese prompt words, covering various style directions such as ink wash, Chinese trendy, realistic, and animation, and is suitable for directly describing the content, atmosphere, and main features of the image.
AI Image Customization
It offers stylized templates such as "My Childhood" and "Magic Academy". Upload a portrait to generate works in the corresponding style, with simple operation steps, making it suitable for non-professional users.
Work Square & One-Click Same Style
Collects users' public works, allows viewing of prompt words and replication for generation, making it easy to quickly gain inspiration and intuitively understand the differences between various styles.
Multi-style Model Library
Built-in with multiple style models, it supports switching art styles for comparison under the same prompt, making it convenient to select the visual solution that best fits the expression goal.
Social Scene Adaptation
Targeted at content consumption for short videos and social platforms, the image generation style and visual form are designed to be easy to spread and share, making it suitable for creating avatars and accompanying images.
AI creation workflow
It forms a closed loop around creation, display and reproduction. Users can not only generate images on their own, but also find inspiration from community works before making secondary adjustments and modifications.
How to use Ketu
- The image customization functions involve facial information.
- Please confirm the platform's data processing instructions and obtain the explicit consent of the person being photographed before uploading.
- After using the "one-click same style" feature, it is recommended to rewrite the prompts to avoid being highly similar to others' works, and it will also be easier to gradually develop your own stable and controllable style preferences.
- When generating portraits for public dissemination or commercial use, please pay attention to portrait rights and the platform's terms of service, and avoid direct use in commercial scenarios such as endorsements.
- First browse the Works Square to confirm your style preferences before starting official creation, which can significantly reduce the number of repeated trials and the overall time cost.
Traffic & popularity
Data month Sep 2026
Ketu receives about 9.1K visits per month.
Monthly visits
12 monthsTraffic sources
- Direct54.9%
- Search22.4%
- Referrals18.5%
- Social4.2%
- AI assistants0.1%
Top countries
- China69.8%
- Germany10.9%
- Mexico9.3%
- Indonesia8.0%
- Singapore1.9%
Estimated from third-party traffic data.
Similar tools
Miaohui AIMiaohui AI is a free AI painting generation platform under Shanghai Yuanxiang Information Technology. Officially, it supports text-to-image, image-to-image and AI video generation, covering styles such as character portraits, landscape photography, product display and artistic illustrations, and provides e-commerce image and prompt word assistance functions.
WHEEThe AI visual creation platform launched by Meitu Inc. supports text-to-image, image-to-image, as well as functions such as image modification, image expansion, object removal, and HD restoration. It can also convert images into videos, targeting individual creators and commercial design scenarios.
Keling AIKuaishou has launched an AI video and image generation platform, featuring native 4K video, motion control, keyframe and audio-video synchronization capabilities, supporting ultra-long video shot segmentation and opening up developer APIs.
SenseMirageThe AI painting platform launched by SenseTime supports Chinese prompt-based text-to-image and image-to-image generation, offers a variety of style models, and supports LoRA training for customizing images and painting styles. Officially, it is available for free and targeted at illustration and IP creation scenarios.
MidjourneyAn AI image generator known for aesthetics, built by an independent, community-funded research lab. Strong stylization, reference images, and character consistency. Subscription-based.
Tencent HunyuanTencent's self-developed multimodal large model product system provides capabilities such as Hunyuan Image Generation in the image direction. It supports Chinese prompt understanding and controllable image generation, can be accessed through open APIs and cloud services, and is targeted at scenarios such as content creation, marketing materials, and brand design.