en.Wedoany.com Reported - Recently, Chinese AI video generation company Sand.ai completed two consecutive rounds of financing, with total funding exceeding $100 million. Investors in this round include Look Capital, Lollapalooza Capital, Ubiquant Venture Capital, Matrix Partners China, Capital Today, Sinovation Ventures, Shunwei Capital, Source Code Capital, Zhongke Chuangxing, Hongtai Fund, Hina Group Venture, Huaye Tiancheng, Yunhui Capital, IDG, and Baidu Ventures, with Xinghan Capital serving as financial advisor.
Sand.ai focuses on AI video generation large models, with its core business revolving around video generation, video editing, multimodal content production, and related AI applications. As demand grows for text-to-video, image-to-video, video continuation, camera control, and long-form video generation, AI video models are becoming a key track in the competitive landscape of generative artificial intelligence.
Public information shows that Sand.ai previously launched the video generation model MAGI-1. This model generates videos using an autoregressive approach, achieving video generation by predicting sequences of continuous video segments, and emphasizes temporal consistency, controllable generation, and long-context processing capabilities. Related technical papers indicate that the largest version of MAGI-1 has 24 billion parameters, supports a maximum context length of 4 million tokens, and offers generation capabilities for image-to-video tasks.
Following this financing, Sand.ai plans to continue advancing the development of next-generation video generation models. Multiple current news pages mention that the company plans to open-source a new generation video generation model in the third quarter of 2026. This model will adopt a MoE architecture, aiming to balance inference efficiency, parameter scale, and the open-source ecosystem. For AI video companies, open-sourcing models helps expand the developer base and may also enhance the model's application coverage in creative tools, advertising and marketing, film and television assistance, and short video production.
The intensive influx of capital into the AI video generation field is primarily related to the production costs of video content and improvements in model capabilities. Compared to text and image generation, video generation requires simultaneous handling of temporal continuity, camera movement, character actions, visual consistency, and physical logic, resulting in higher training costs, inference costs, and engineering complexity. Teams with comprehensive capabilities in pre-training, post-training, and infrastructure engineering are more likely to attract attention from investment institutions.
Public information from Source Code Capital indicates that it led Sand.ai's seed round in 2024, making it one of the company's early institutional investors, and it has continued to invest in subsequent rounds. Related statements from Matrix Partners China also mention that Sand.ai possesses capabilities in pre-training, post-training, and Infra engineering, and has invested early in the direction of multimodal autoregression and MoE architecture. This suggests that investors are not only focused on single product traffic but also on underlying model roadmaps and engineering capabilities.
However, the commercialization of AI video generation remains in a rapidly changing phase. Model effectiveness, generation costs, copyright compliance, content safety, video authenticity recognition, and user payment conversion will all impact the company's future development. Whether Sand.ai can translate its model capabilities into stable product revenue after this substantial financing will depend on the performance of its open-source models, commercial product experience, developer ecosystem, and enterprise-level customer expansion.
Key points for subsequent observation will focus on the release timeline of the new generation MoE video model, the scope of open-sourcing, model parameter scale, inference costs, commercial product updates, and overseas market performance. If Sand.ai can achieve breakthroughs in video generation quality, generation speed, and cost control, it will continue to be one of the important entrepreneurial companies in China's AI video generation track.
This article is compiled by Wedoany. All AI citations must indicate the source as "Wedoany". If there is any infringement or other issues, please notify us promptly, and we will modify or delete it accordingly. Email: news@wedoany.com
