Alibaba Group's Tongyi announced that its latest developed AI video generation large model - Tongyi Wanxiang AI Video Generation, is officially online. This technology, with its powerful visual dynamic generation capability, can quickly generate video content with various artistic styles and cinematic quality based on text prompts.

Today, ByteDance's Volcano Engine announced that the Doubao Large Model will release a video generation model on September 24, bringing upgrades to more capabilities in the model family. It is understood that the Doubao Large Model was officially launched at the Volcano Engine Original Power Conference on May 15, 2024.

In today's rapidly advancing field of artificial intelligence, AI video generation technology has reached a significant milestone. Runway from New York and Luma AI from San Francisco have almost simultaneously launched their respective video generation APIs to the market, providing developers and businesses with unprecedented creative tools and attracting widespread attention in the industry. Runway's Gen-3Alpha Turbo AI video model API was the first to be unveiled. This product is currently in a phased rollout, initially accessible only to select partners.

Alibaba's Tongyi is set to unveil its video generation large model at the upcoming Yunqi Conference, attracting significant attention from the industry. Currently, the Tongyi App channel and Tongyi Wanxiang PC version have quietly launched the 'Video Generation' entry, although the functionality is not yet open and users can only make reservations. According to page information, this new video generation large model is independently developed by Alibaba's Tongyi Laboratory, featuring two main creation modes: 'Text-to-Video' and 'Image-to-Video', allowing users to generate corresponding video content from text or images. This launch will undoubtedly bring new opportunities for content creators.

After MiniMax launched its first AI HD video generation model, its subsidiary Conch AI has added the features of 'AI Music Creation' and 'AI Video Generation'. Users can automatically generate music by creating lyrics and selecting music styles, while videos can be generated simply by inputting creative descriptions. The music feature offers a variety of styles, but the quality of the generated music is average; the video feature can automatically create content based on input descriptions, presenting a poetic effect, although the image quality still has room for improvement. Currently, these two features are available for users to experience at no cost for a limited time.

Recently, MiniMax secretly launched its first video generation large model and showcased a 2-minute video titled 'Magic Coin' generated by this model. Founder Yan Junjie revealed that the model outperforms Runway in video generation quality, but specific parameters and technical details have not yet been disclosed. MiniMax plans to continue iterating in future versions to enhance data, algorithms, and usage details, and aims to add the ability to generate videos from images and a combination of text and images in future updates. Regarding commercialization, the company will consider it once the new version reaches a satisfactory state.

The domestic open source video generation model CogVideoX-5B has been officially released in the Mota ModelScope community, significantly improving the quality and visual effects of video generation. Based on the large-scale DiT model, this model utilizes a 3D causal variational autoencoder and expert Transformer technology, achieving spatio-temporal joint modeling through 3D-RoPE positional encoding and a 3D full attention mechanism. The use of progressive training techniques allows the model to generate long videos with distinct motion features, coherence, and high quality.

Since the launch of the Kuaishou 'Keling' AI web platform on July 6, it has gained considerable popularity among creators, generating over 16 million videos. As of August 27, 2024, at the Guanghe Creator Conference, over 1.6 million users have utilized this AI model, demonstrating its immense potential. The 'Keling' AI can generate high-definition videos up to 2 minutes long, supporting free aspect ratio adjustments, providing creators with creative freedom. On July 30, the AI officially launched its membership system globally, becoming the world's first large model for generating true image-level videos.

Aishi Technology announced the global launch of version V2.5 of its video generation product PixVerse, which has achieved significant improvements in video generation speed, dynamic performance, and picture quality. The new version has particularly optimized the model, enhancing prompt understanding and the aesthetic level of video generation, increasing the video generation speed by 200%, while supporting output up to 4K resolution to ensure clarity in video details.

In July this year, Aishi Technology released the newly upgraded AI video generation product PixVerse V2, which is based on the Diffusion+Transformer (DiT) architecture, achieving multiple innovations in video generation technology. PixVerse V2 offers longer, more consistent, and more engaging video generation capabilities, supporting the generation of multiple video clips at once.

Musk sparked widespread attention by releasing a video of himself dancing with Trump, utilizing an AI video generation tool called Viggle AI, which led to over 130 million views. The magic of Viggle AI lies in its templated operation; users simply upload photos and choose action templates to create smooth, natural dance videos. The tool's multi-character control feature, Multi, further ignited users' creativity, allowing them to produce various creative videos that spread online, bringing a substantial user base to Viggle AI.

Runway ML has launched the AI video generation model Gen-3 Alpha Turbo, which offers a 7-fold increase in speed and a 50% reduction in costs compared to its predecessor, making AI video production more accessible. Turbo is suitable for all plans, including free trials, allowing users to enjoy an efficient video generation experience without incurring additional costs. The generation speed of Gen-3 Alpha Turbo is so fast that it surpasses typing speed, effectively addressing the time lag issue of AI video generation models. In terms of pricing, video costs only $5 per second.

An AI-generated video featuring Turkish shooter Devrim Değirmenci has gone viral on social media, catching the attention of tech mogul Elon Musk. Devrim's relaxed demeanor at the Olympics has made him a hot topic online, with AI creators utilizing cutting-edge technology to transform his competition photos into imaginative videos that showcase the contrast between intense competition and his laid-back daily life, infused with humor, instantly drawing massive attention. The video has amassed over 170 million views, highlighting the influence of AI creations on social media. This phenomenon not only reflects the public's interest in AI creation.

Zhipu AI announces the open source of its video generation model CogVideoX, aiming to promote the rapid development and application of video generation technology. The CogVideoX model is based on advanced large model technology and is capable of meeting commercial-grade application requirements. The currently open sourced CogVideoX-2B version requires only 18GB of GPU memory for inference at FP-16 precision and 40GB for fine-tuning, allowing inference with a single 4090 GPU and fine-tuning with a single A6000 GPU.

With the official launch of the global membership service, the official website shows that the membership system is divided into three levels, similar to the domestic setup. For example, for the monthly card, the prices for the three membership levels are $10, $37, and $92, which correspond to 660, 3000, and 8000 inspiration points, respectively. These inspiration points can be used to generate 66, 300, or 800 five-second high-quality videos. In addition to the monthly card, various package options such as quarterly, semi-annual, and annual cards are also available.

Baidu Intelligent Cloud and Shengshu Technology have officially launched the video model Vidu, bringing new possibilities to the creative field. Vidu is the first fully self-researched video model in China, with performance fully comparable to OpenAI's Sora. It is aimed at creative fields such as art design, game production, film and animation, and social entertainment, and is dedicated to stimulating the productivity and creativity of all humanity.

Today, Shenshu Technology announced the global launch of its AI video generation model, Vidu. Vidu is a model that utilizes artificial intelligence technology to generate videos from text or images. In April of this year, Shenshu Technology partnered with Tsinghua University to release China's first large video model - 'Vidu', marking an important step for China in video generation technology.

Zhipu AI has launched CogVideoX, a next-generation video generation model that enhances temporal frame relationship capturing through a 3D VAE structure and 3D RoPE module. This model integrates text, time, and space into a transformer architecture, allowing for efficient video generation that meets user needs. It is available on the Zhipu Qingyan platform for free video generation services, featuring fast generation, effective instruction adhere....

During its OpenDay event, Zhipu AI introduced an AI large model named Qingying, capable of generating videos from any text. Qingying can produce high-definition videos at a resolution of 1440x960 within 30 seconds. Users need only to input a text (i.e., Prompt) and choose a desired style, such as cartoon 3D, black and white, oil painting, cinematic, etc., to generate the video. Qingying is now available on the Qingyan App, accessible to all users.