10 comic dramas have racked up a total of 1.2 billion views! The full-link AI platform of Zizi Animation is set to build the "industrialized assembly line" for the comic drama industry.
Generative AI is lowering the barriers to producing images and videos, but for professional content creators, there is still a clear gap between "being able to generate a video clip" and "being able to continuously produce complete works".
Research Report on the Development of China's Online Audio-Visual Industry (2025) shows that as of December 2024, the number of short video users in China reached 1.04 billion, with a usage rate of 93.8%; the White Paper on the Development of China's Micro Drama Industry (2025) indicates that the number of micro drama users in China has reached 696 million. While the user base is growing, industry competition has gradually shifted from the pursuit of output and traffic to competition over content quality, production stability and continuous supply capacity.
As a new content form formed by the combination of micro dramas and generative AI, AI manhua dramas are also entering a stage of rapid development. According to calculations by iiMedia Research, the market size of Chinese animated micro dramas reached 189.8 billion yuan in 2025, and is expected to exceed 850 billion yuan by 2030. Market expansion has brought greater content demand, and also put forward higher requirements for creators' professional production capabilities.
Traditional manhua drama production involves multiple links including creative conception, script adaptation, character design, scene setting, storyboard production, image generation, video synthesis, editing, dubbing and distribution, which usually requires the collaborative work of screenwriters, artists, directors and post-production staff. Although existing AI image and video tools can reduce the operating costs of some links, most products still focus on generating a single image or a single video clip. Creators often need to transfer materials, prompts and parameters between multiple tools, and it is difficult to stably control character consistency, scene continuity, shot transition and overall style.
The node-based workflow represented by ComfyUI provides a high degree of freedom, but at the same time requires users to have technical capabilities such as model deployment, node configuration and parameter debugging. For individual creators and small and medium-sized production teams, how to retain the control capabilities required for professional production while lowering the technical threshold has become a key issue for AI manhua dramas to move towards large-scale production.
Shifting from single-point generation to full-link production
Zizi Animation was launched in 2023. The project is not positioned to provide a single image or video generation capability, but to build a production system covering content creation, production and manufacturing, computing power supply, copyright acquisition and multi-platform distribution around AI manhua dramas.
Zizi Animation has currently formed a product matrix consisting of AI script creation tools, manhua drama production tools, OmniBoard infinite canvas, elastic computing power platform and distribution platform. Different products undertake the links of creative conception, script generation, character and scene design, storyboard production, animation generation, video synthesis and work distribution respectively, reducing the problem of creators repeatedly switching between multiple independent tools.
Among them, OmniBoard adopts an infinite canvas and visual workflow, organizing text, images, videos, character assets and generation nodes in the same creation space. Creators can freely build workflows according to project needs, and adjust key characters, scenes and shots through multi-view editing and manual intervention. OmniBoard also supports multi-person collaboration, enabling team members to cooperate in real time around script structure, character setting and production progress.
In terms of production methods, Zizi Animation does not equate AI automation with the complete elimination of manual work. Zizi Animation assigns highly repetitive tasks of character, scene, storyboard and video generation to AI for processing, while retaining manual operations such as key shot retouching, narrative rhythm adjustment and finished work quality control. The resulting model of "coexistence of AI mass production and manual professional control" mainly solves the problem of difficulty in balancing large-scale efficiency and finished work quality.
Facing the industry status quo that video models are updated rapidly and the capabilities of different models are significantly different, Zizi Animation has built a multi-model access system to decouple the underlying models from the upper-level creation process. The project team can access corresponding models according to different production needs such as image, video, action and style, while reducing the impact of model update or replacement on the original workflow.
On this basis, Zizi Animation has opened up the image and video plug-in system, allowing creators to expand the workflow according to model capabilities, theme types and project requirements. According to data provided by the project party, platform users have cumulatively created and used tens of thousands of custom plug-ins. Plug-ins are not only single-function extensions, but also carry the parameter combinations, production methods and process experience accumulated by creators in real projects, enabling the platform's capabilities to continue to expand with user usage.
Video generation has high requirements for the stability of computing power supply. The computing requirements of different models, task concurrency and resource prices may fluctuate, which directly affects the production cycle of works. To this end, Zizi Animation has independently developed a multi-channel elastic GPU scheduling system, which allocates computing power according to model type, task scale and real-time resource situation, so as to improve resource utilization efficiency and the stability of batch tasks. The underlying computing power capabilities can also provide services to external teams through APIs to meet the mass production needs of studios and enterprise users.
Extending from content production to copyright and distribution
Completing video generation does not mean that the content production link has ended. For professional manhua drama creators, copyright acquisition, content review, multi-platform distribution, data tracking and subsequent operation also require a lot of time investment. Single-point generation tools usually do not participate in these links, and creators still need to find copyright resources by themselves, understand the rules of different platforms and complete work release.
Therefore, Zizi Animation extends its product capabilities from the production end to the content and distribution end. The project team has built channels for authorized content acquisition and copyright application, connected with content platforms or copyright resource parties such as Qimao, China Literature and Fanqie Novel, to provide creators with bibliographic selection, authorization application and adaptation process support. Relevant content completes copyright confirmation before entering the production link, making the material source, authorization process and distribution link traceable.
In the distribution link, the Zizi Animation distribution platform provides work submission, multi-platform distribution, play data monitoring and content operation support. The connection between the production end and the distribution end enables creators to adjust theme selection, content rhythm and production methods according to the communication performance of works after they go online. Compared with tools that only provide generation capabilities, this connection from content source, work production to distribution feedback requires the accumulation of technical systems, platform rules, copyright resources and practical operation experience at the same time.
Zizi Animation is currently mainly oriented to professional AI manhua drama creators, individual producers, small and medium-sized studios and MCN institutions, and also provides corresponding tools and services for teaching teams and enterprise customers. Zizi Animation's business model uses free creation tools as the user entry, and value-added services include SaaS membership, computing power services, API capabilities, distribution services, as well as enterprise version authorization and customized services for institutions. Different services correspond to the needs of creators in advanced functions, mass production, computing power invocation, work distribution and team management.
Compared with relying only on single tool subscriptions, Zizi Animation's product structure covers multiple links for creators from starting production to completing distribution. The core of this model is not simply adding paid items, but improving the platform's coverage of the professional production process, so that tool capabilities, computing power infrastructure and content distribution resources can be connected to each other.
User and work data verify professional production demands
Liu Ziyao, the founder of Zizi Animation, worked at ByteDance from 2018 to 2020, and engaged in short video content ecology related work at Tencent WeChat Channels from 2021 to 2023. The core project team currently consists of 10 full-time members, adopting a full-stack team model to advance different product lines in parallel. The Zizi Animation product was officially launched in October 2025, and then gradually improved its capabilities related to computing power, creation, distribution and content.
According to the data provided by the project party as of August 2026, Zizi Animation platform has accumulated more than 120,000 users, precipitated more than 40,000 community users, established more than 100 creator communities, and nearly 1,000 creators have completed manhua drama distribution through relevant services. These users do not only stay in the single generation link, but further enter professional processes such as script processing, character and scene production, batch generation and work distribution.
At the work end, the 10 representative manhua dramas disclosed by the project party have accumulated more than 1.2 billion views. Among them, works such as Lifelong Brothers, Half-Life Affection, Don't Blame Me for Being Cruel, Who Made You Touch My 300,000 Fry, Who Has the Final Say on My Parking Space and Cold Noodle Shop have all been produced and distributed through relevant products of Zizi Animation. These cases provide practical project verification for the platform's capabilities in long-form content production, character continuity, batch shot production and multi-platform distribution.
From the perspective of competition logic, AI video models are still iterating rapidly, and the single-generation effect is easy to be caught up with the progress of underlying models. The more long-term barriers of AI manhua drama production platforms come from the joint accumulation of professional workflows, multi-model adaptation, computing power scheduling, user plug-in ecology, authorized content resources and distribution networks.
What Zizi Animation tries to build is not a short-term advantage in a certain generation link, but a set of production systems that can undertake real content projects. As AI manhua dramas shift from individual attempts to team-based, large-scale and high-quality production, industry competition will gradually shift from "whether content can be generated" to "whether works can be stably completed and enter the distribution link". The full-link capability built around this change constitutes Zizi Animation's differentiated direction at the current stage.