{"id":101583,"date":"2025-02-21T13:55:40","date_gmt":"2025-02-21T13:55:40","guid":{"rendered":"https:\/\/peraltafinancing.com\/analytics\/chinas-new-ai-video-star-step-video-t2v\/"},"modified":"2025-02-21T13:55:40","modified_gmt":"2025-02-21T13:55:40","slug":"chinas-new-ai-video-star-step-video-t2v","status":"publish","type":"post","link":"https:\/\/fivemor.com\/?p=101583","title":{"rendered":"China\u2019s New AI Video Star: Step-Video-T2V"},"content":{"rendered":"<p> <br \/>\n<\/p>\n<div id=\"article-start\">\n<p>China is advancing rapidly in generative AI, building on successes like <a href=\"https:\/\/www.analyticsvidhya.com\/blog\/2025\/02\/goku-ai\/\" target=\"_blank\" rel=\"noreferrer noopener\">DeepSeek<\/a> models and <a href=\"https:\/\/www.analyticsvidhya.com\/blog\/2025\/01\/kimi-k1-5\/\" target=\"_blank\" rel=\"noreferrer noopener\">Kimi k1.5<\/a> in language models. Now, it\u2019s leading the vision domain with <a href=\"https:\/\/www.analyticsvidhya.com\/blog\/2025\/02\/omnihuman\/\" target=\"_blank\" rel=\"noreferrer noopener\">OmniHuman<\/a> and <a href=\"https:\/\/www.analyticsvidhya.com\/blog\/2025\/02\/goku-ai\/\">Goku<\/a> excelling in 3D modeling and video synthesis. With Step-Video-T2V, China directly challenges top text-to-video models like <a href=\"https:\/\/www.analyticsvidhya.com\/blog\/2024\/12\/openai-sora\/\" target=\"_blank\" rel=\"noreferrer noopener\">Sora<\/a>, <a href=\"https:\/\/www.analyticsvidhya.com\/blog\/2024\/12\/googles-veo-2\/\">Veo 2<\/a>, and Movie Gen. Developed by Stepfun AI, Step-Video-T2V is a 30B-parameter model that generates high-quality, 204-frame videos. It leverages a Video-VAE, bilingual encoders, and a 3D-attention DiT to set a new video generation standard. Does it address text-to-video\u2019s core challenges? Let\u2019s dive in.<\/p>\n<h2 class=\"wp-block-heading\" id=\"h-challenges-in-text-to-video-models\">Challenges in Text-to-Video Models<\/h2>\n<p>While text-to-video models have come a long way, they still face fundamental hurdles:<\/p>\n<ul class=\"wp-block-list\">\n<li><strong>Complex Action Sequences<\/strong> \u2013 Current models struggle to generate realistic videos that follow intricate action sequences, such as a gymnast performing flips or a basketball bouncing realistically.<\/li>\n<li><strong>Physics and Causality<\/strong> \u2013 Most diffusion-based models fail to simulate the real world effectively. Object interactions, gravity, and physical laws are often overlooked.<\/li>\n<li><strong>Instruction Following<\/strong> \u2013 Models frequently miss key details in user prompts, especially when dealing with rare concepts (e.g., a penguin and an elephant in the same video).<\/li>\n<li><strong>Computational Costs<\/strong> \u2013 Generating high-resolution, long-duration videos is <strong>extremely resource-intensive<\/strong>, limiting accessibility for researchers and creators.<\/li>\n<li><strong>Captioning and Alignment<\/strong> \u2013 Video models rely on massive datasets, but poor video captioning results in weak prompt adherence, leading to <strong>hallucinated content<\/strong>.<\/li>\n<\/ul>\n<h2 class=\"wp-block-heading\" id=\"h-how-step-video-t2v-is-solving-these-problems\">How Step-Video-T2V is Solving These Problems?<\/h2>\n<p>Step-Video-T2V tackles these challenges with <strong>several innovations<\/strong>:<\/p>\n<ul class=\"wp-block-list\">\n<li><strong>Deep Compression Video-VAE<\/strong>: Achieves <strong>16\u00d716 spatial and 8x temporal compression<\/strong>, significantly reducing computational requirements while maintaining high video quality.<\/li>\n<li><strong>Bilingual Text Encoders<\/strong>: Integrates <strong>Hunyuan-CLIP and Step-LLM<\/strong>, allowing the model to process prompts effectively in both <strong>Chinese and English<\/strong>.<\/li>\n<li><strong>3D Full-Attention DiT<\/strong>: Instead of traditional spatial-temporal attention, this approach enhances <strong>motion continuity and scene consistency<\/strong>.<\/li>\n<li><strong>Video-DPO (Direct Preference Optimization)<\/strong>: Incorporates <strong>human feedback loops<\/strong> to reduce artifacts, improve realism, and align generated content with user expectations.<\/li>\n<\/ul>\n<h2 class=\"wp-block-heading\" id=\"h-model-architecture\">Model Architecture<\/h2>\n<p>The Step-Video-T2V model architecture is structured around a three-part pipeline to effectively process text prompts and generate high-quality videos. The model integrates a bilingual text encoder, a Variational Autoencoder (Video-VAE), and a Diffusion Transformer (DiT) with 3D Attention, setting it apart from traditional text-to-video models.<\/p>\n<h3 class=\"wp-block-heading\" id=\"h-1-text-encoding-with-bilingual-understanding\">1. Text Encoding with Bilingual Understanding<\/h3>\n<p>At the input stage, Step-Video-T2V employs <strong>two powerful bilingual text encoders:<\/strong><\/p>\n<ul class=\"wp-block-list\">\n<li><strong>Hunyuan-CLIP<\/strong>: A vision-language model optimized for <strong>semantic alignment<\/strong> between text and images.<\/li>\n<li><strong>Step-LLM<\/strong>: A large language model specialized in <strong>understanding complex instructions<\/strong> in both <strong>Chinese and English<\/strong>.<\/li>\n<\/ul>\n<p>These encoders process the <strong>user prompt<\/strong> and convert it into a meaningful <strong>latent representation<\/strong>, ensuring that the model accurately follows instructions.<\/p>\n<h3 class=\"wp-block-heading\" id=\"h-2-variational-autoencoder-video-vae-for-compression\">2. Variational Autoencoder (Video-VAE) for Compression<\/h3>\n<p>Generating long, high-resolution videos is computationally expensive. Step-Video-T2V tackles this issue with a <strong>deep compression Variational Autoencoder (Video-VAE)<\/strong> that reduces video data efficiently:<\/p>\n<ul class=\"wp-block-list\">\n<li><strong>Spatial compression (16\u00d716)<\/strong> and <strong>temporal compression (8x)<\/strong> reduce video size while preserving motion details.<\/li>\n<li>This enables <strong>longer sequences (204 frames)<\/strong> with <strong>lower compute costs<\/strong> than previous models.<\/li>\n<\/ul>\n<h3 class=\"wp-block-heading\" id=\"h-3-diffusion-transformer-dit-with-3d-full-attention\">3. Diffusion Transformer (DiT) with 3D Full Attention<\/h3>\n<p>The core of Step-Video-T2V is its <strong>Diffusion Transformer (DiT) with 3D Full Attention<\/strong>, which significantly improves motion smoothness and scene coherence.\u00a0<\/p>\n<p>The <strong>ith block<\/strong> of the DiT consists of multiple components that refine the video generation process:<\/p>\n<h4 class=\"wp-block-heading\" id=\"h-key-components-of-each-transformer-block\"><strong>Key Components of Each Transformer Block<\/strong><\/h4>\n<ul class=\"wp-block-list\">\n<li><strong>Cross-Attention<\/strong>: Ensures <strong>better text-to-video alignment<\/strong> by conditioning the generated frames on the text embedding.<\/li>\n<li><strong>Self-Attention (with RoPE-3D)<\/strong>: Uses <strong>Rotary Positional Encoding (RoPE-3D)<\/strong> to enhance <strong>spatial-temporal understanding<\/strong>, ensuring that objects move naturally across frames.<\/li>\n<li><strong>QK-Norm (Query-Key Normalization)<\/strong>: Improves the stability of attention mechanisms, reducing inconsistencies in object positioning.<\/li>\n<li><strong>Gate Mechanisms<\/strong>: These <strong>adaptive gates<\/strong> regulate information flow, preventing <strong>overfitting to specific patterns<\/strong> and improving generalization.<\/li>\n<li><strong>Scale\/Shift Operations<\/strong>: Normalize and fine-tune intermediate representations, ensuring smooth transitions between video frames.<\/li>\n<\/ul>\n<h3 class=\"wp-block-heading\" id=\"h-4-adaptive-layer-normalization-adaln-single\">4. Adaptive Layer Normalization (AdaLN-Single)<\/h3>\n<ul class=\"wp-block-list\">\n<li>The model also includes <strong>Adaptive Layer Normalization (AdaLN-Single)<\/strong>, which adjusts activations dynamically based on the <strong>timestep (t)<\/strong>.<\/li>\n<li>This ensures <strong>temporal consistency<\/strong> across the video sequence.<\/li>\n<\/ul>\n<h2 class=\"wp-block-heading\" id=\"h-how-does-step-video-t2v-work\">How Does Step-Video-T2V Work?<\/h2>\n<p>The <strong>Step-Video-T2V<\/strong> model is a cutting-edge <strong>text-to-video AI system<\/strong> that generates high-quality motion-rich videos based on textual descriptions. The working mechanism involves multiple sophisticated AI techniques to ensure smooth motion, adherence to prompts, and realistic output. Let\u2019s break it down step by step:<\/p>\n<h3 class=\"wp-block-heading\" id=\"h-1-user-input-text-encoding\">1. User Input (Text Encoding)<\/h3>\n<ul class=\"wp-block-list\">\n<li>The model starts by <strong>processing user input<\/strong>, which is a text prompt describing the desired video.<\/li>\n<li>This is done using <strong>bilingual text encoders<\/strong> (e.g., <strong>Hunyuan-CLIP and Step-LLM<\/strong>).<\/li>\n<li>The <strong>bilingual capability<\/strong> ensures that prompts in <strong>both English and Chinese<\/strong> can be understood accurately.<\/li>\n<\/ul>\n<h3 class=\"wp-block-heading\" id=\"h-2-latent-representation-compression-with-video-vae\">2. Latent Representation (Compression with Video-VAE)<\/h3>\n<ul class=\"wp-block-list\">\n<li>Video generation is computationally heavy, so the model employs a <strong>Variational Autoencoder (VAE)<\/strong> specialized for video compression, called <strong>Video-VAE<\/strong>.<\/li>\n<li><strong>Function of Video-VAE:<\/strong>\n<ul class=\"wp-block-list\">\n<li>Compresses video frames into a <strong>lower-dimensional latent space<\/strong>, significantly reducing <strong>computational costs<\/strong>.<\/li>\n<li><strong>Maintains key video quality aspects<\/strong>, such as <strong>motion continuity, textures, and object details<\/strong>.<\/li>\n<li>Uses a <strong>16\u00d716 spatial and 8x temporal compression<\/strong>, making the model efficient while preserving high fidelity.<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n<h3 class=\"wp-block-heading\" id=\"h-3-denoising-process-diffusion-transformer-with-3d-full-attention\">3. Denoising Process (Diffusion Transformer with 3D Full Attention)<\/h3>\n<ul class=\"wp-block-list\">\n<li>After obtaining the latent representation, the next step is the <strong>denoising process<\/strong>, which refines the video frames.<\/li>\n<li>This is done using a <strong>Diffusion Transformer (DiT)<\/strong>, an advanced model designed for generating highly realistic videos.<\/li>\n<li><strong>Key innovation:<\/strong>\n<ul class=\"wp-block-list\">\n<li>The <strong>Diffusion Transformer<\/strong> applies <strong>3D Full Attention<\/strong>, a powerful mechanism that focuses on <strong>spatial, temporal, and motion dynamics<\/strong>.<\/li>\n<li>The use of <strong>Flow Matching<\/strong> helps <strong>enhance the movement consistency<\/strong> across frames, ensuring smoother video transitions.<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n<h3 class=\"wp-block-heading\" id=\"h-4-optimization-fine-tuning-and-video-dpo-training\">4. Optimization (Fine-Tuning and Video-DPO Training)<\/h3>\n<p>The generated video undergoes an optimization phase, making it more <strong>accurate, coherent, and visually appealing<\/strong>. This involves:<\/p>\n<ul class=\"wp-block-list\">\n<li><strong>Fine-tuning the model<\/strong> with high-quality data to improve its ability to follow complex prompts.<\/li>\n<li><strong>Video-DPO (Direct Preference Optimization)<\/strong> training, which incorporates <strong>human feedback<\/strong> to:\n<ul class=\"wp-block-list\">\n<li>Reduce unwanted artifacts.<\/li>\n<li>Improve realism in motion and textures.<\/li>\n<li>Align video generation with user expectations.<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n<h3 class=\"wp-block-heading\" id=\"h-5-final-output-high-quality-204-frame-video\">5. Final Output (High-Quality 204-Frame Video)<\/h3>\n<ul class=\"wp-block-list\">\n<li>The final video is <strong>204 frames long<\/strong>, meaning it provides a <strong>significant duration for storytelling<\/strong>.<\/li>\n<li><strong>High-resolution generation<\/strong> ensures crisp visuals and clear object rendering.<\/li>\n<li><strong>Strong motion realism<\/strong> means the video maintains <strong>smooth and natural movement<\/strong>, making it suitable for complex scenes like human gestures, object interactions, and dynamic backgrounds.<\/li>\n<\/ul>\n<h2 class=\"wp-block-heading\" id=\"h-benchmarking-against-competitors\">Benchmarking Against Competitors<\/h2>\n<p>Step-Video-T2V is evaluated on <strong>Step-Video-T2V-Eval<\/strong>, a <strong>128-prompt benchmark<\/strong> covering <strong>sports, food, scenery, surrealism, people, and animation<\/strong>. Compared against leading models, it delivers <strong>state-of-the-art performance in motion dynamics and realism.<\/strong><\/p>\n<ol class=\"wp-block-list\">\n<li><strong>Outperforms HunyuanVideo<\/strong> in overall video quality and smoothness.<\/li>\n<li><strong>Rivals Movie Gen Video<\/strong> but lags in fine-grained aesthetics due to limited high-quality labeled data.<\/li>\n<li><strong>Beats Runway Gen-3 Alpha<\/strong> in motion consistency but slightly lags in cinematic appeal.<\/li>\n<li><strong>Challenges Top Chinese commercial models (T2VTopA and T2VTopB)<\/strong> but falls short in aesthetic quality due to lower resolution (540P vs. 1080P).<\/li>\n<\/ol>\n<h4 class=\"wp-block-heading\" id=\"h-performance-metrics\">Performance Metrics<\/h4>\n<p>Step-Video-T2V introduces <strong>new evaluation criteria<\/strong>:<\/p>\n<ul class=\"wp-block-list\">\n<li><strong>Instruction Following<\/strong> \u2013 Measures how well the generated video aligns with the prompt.<\/li>\n<li><strong>Motion Smoothness<\/strong> \u2013 Rates the natural flow of actions in the video.<\/li>\n<li><strong>Physical Plausibility<\/strong> \u2013 Evaluates whether movements follow the laws of physics.<\/li>\n<li><strong>Aesthetic Appeal<\/strong> \u2013 Judges the artistic and visual quality of the video.<\/li>\n<\/ul>\n<p>In human evaluations, <strong>Step-Video-T2V consistently outperforms competitors in motion smoothness and physical plausibility<\/strong>, making it one of the most advanced open-source models.<\/p>\n<h2 class=\"wp-block-heading\" id=\"h-how-to-access-step-video-t2v\">How to Access Step-Video-T2V?<\/h2>\n<p><strong>Step 1: <\/strong>Visit the official website <a href=\"https:\/\/yuewen.cn\/videos\" target=\"_blank\" rel=\"noreferrer noopener nofollow\">here<\/a>.<\/p>\n<p><strong>Step 2: <\/strong>Sign up using your mobile number.<\/p>\n<p><strong>Note:<\/strong> Currently, registrations are open only for a limited number of countries. Unfortunately, it is not available in India, so I couldn\u2019t sign up. However, you can try if you\u2019re located in a supported region.<\/p>\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-full is-resized\"><img loading=\"lazy\" decoding=\"async\" width=\"375\" height=\"472\" src=\"https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/Login.webp\" alt=\"\" class=\"wp-image-222786\" style=\"width:303px;height:auto\" srcset=\"https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/Login.webp 375w, https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/Login-238x300.webp 238w, https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/Login-150x189.webp 150w\" sizes=\"auto, (max-width: 375px) 100vw, 375px\"\/><\/figure>\n<\/div>\n<p><strong>Step 3: <\/strong>Add in your prompt and start generating amazing videos!<\/p>\n<figure class=\"wp-block-image size-full\"><img loading=\"lazy\" decoding=\"async\" width=\"872\" height=\"476\" src=\"https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/Step-Video-T2V.webp\" alt=\"\" class=\"wp-image-222781\" srcset=\"https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/Step-Video-T2V.webp 872w, https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/Step-Video-T2V-300x164.webp 300w, https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/Step-Video-T2V-768x419.webp 768w, https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/02\/Step-Video-T2V-150x82.webp 150w\" sizes=\"auto, (max-width: 872px) 100vw, 872px\"\/><\/figure>\n<h2 class=\"wp-block-heading\" id=\"h-example-of-vidoes-created-by-step-video-t2v\">Example of Vidoes Created by Step-Video-T2V<\/h2>\n<p>Here are some videos generated by this tool. I have taken these from their official site. <\/p>\n<h3 class=\"wp-block-heading\" id=\"h-van-gogh-in-paris\">Van Gogh in Paris<\/h3>\n<p><strong>Prompt:<\/strong> \u201c<em>On the streets of Paris, Van Gogh is sitting outside a cafe, painting a night scene with a drawing board in his hand. The camera is shot in a medium shot, showing his focused expression and fast-moving brush. The street lights and pedestrians in the background are slightly blurred, using a shallow depth of field to highlight his image. As time passes, the sky changes from dusk to night, and the stars gradually appear. The camera slowly pulls away to see the comparison between his finished work and the real night scene.\u201d<\/em><\/p>\n<p>\n  <iframe src=\"https:\/\/stepvideot2v.com\/videos\/stepvideo-t2v-demo-2.mp4\" loading=\"lazy\" title=\"Millennium Falcon Journey\" allowfullscreen=\"\"><\/iframe>\n<\/p>\n<h3 class=\"wp-block-heading\" id=\"h-millennium-falcon-journey\">Millennium Falcon Journey<\/h3>\n<p><strong>Prompt:<\/strong> \u201c<em>In the vast universe, the Millennium Falcon in Star Wars is traveling across the stars. The camera shows the spacecraft flying among the stars in a distant view. The camera quickly follows the trajectory of the spacecraft, showing its high-speed shuttle. Entering the cockpit, the camera focuses on the facial expressions of Han Solo and Chewbacca, who are nervously operating the instruments. The lights on the dashboard flicker, and the background starry sky quickly passes by outside the porthole.\u201d<\/em><\/p>\n<p>\n  <iframe src=\"https:\/\/stepvideot2v.com\/videos\/stepvideo-t2v-demo-3.mp4\" loading=\"lazy\" title=\"Millennium Falcon Journey\" allowfullscreen=\"\"><\/iframe>\n<\/p>\n<h2 class=\"wp-block-heading\" id=\"h-conclusion\">Conclusion<\/h2>\n<p>Step-Video-T2V isn\u2019t available outside China yet. Once it\u2019s public, I\u2019ll test and share my review. Still, it signals a major advance in China\u2019s generative AI, proving its labs are shaping multimodal AI\u2019s future alongside OpenAI and DeepMind. The next step for video generation demands better instruction-following, physics simulation, and richer datasets. Step-Video-T2V paves the way for open-source video models, empowering global researchers and creators. China\u2019s AI momentum suggests more realistic and efficient text-to-video innovations ahead<\/p>\n<div class=\"border-top py-3 author-info my-4\">\n<div class=\"author-card d-flex align-items-center\">\n<div class=\"flex-shrink-0 overflow-hidden\">\n                                    <a href=\"https:\/\/www.analyticsvidhya.com\/blog\/author\/nitika-sharma\/\" class=\"text-decoration-none active-avatar\"><br \/>\n                                                                       <img decoding=\"async\" src=\"https:\/\/av-eks-lekhak.s3.amazonaws.com\/media\/lekhak-profile-images\/converted_image_A027XT6.webp\" width=\"48\" height=\"48\" alt=\"Nitika Sharma\" loading=\"lazy\" class=\"rounded-circle\"\/><\/p>\n<p>                                <\/a>\n                                <\/div>\n<\/p><\/div>\n<p>Hello, I am Nitika, a tech-savvy Content Creator and Marketer. Creativity and learning new things come naturally to me. I have expertise in creating result-driven content strategies. I am well versed in SEO Management, Keyword Operations, Web Content Writing, Communication, Content Strategy, Editing, and Writing.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n\n","protected":false},"excerpt":{"rendered":"<p>China is advancing rapidly in generative AI, building on successes like DeepSeek models and Kimi k1.5 in language models. Now, it\u2019s leading the vision domain with OmniHuman and Goku excelling in 3D modeling and video synthesis. With Step-Video-T2V, China directly challenges top text-to-video models like Sora, Veo 2, and Movie Gen. Developed by Stepfun AI, [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":101584,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[12033],"tags":[26301,817,47111,1236],"dealstore":[],"offerexpiration":[],"class_list":["post-101583","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-analytics","tag-chinas","tag-star","tag-stepvideot2v","tag-video"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v26.4 - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>China\u2019s New AI Video Star: Step-Video-T2V - Som2ny Network<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/fivemor.com\/?p=101583\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"China\u2019s New AI Video Star: Step-Video-T2V - Som2ny Network\" \/>\n<meta property=\"og:description\" content=\"China is advancing rapidly in generative AI, building on successes like DeepSeek models and Kimi k1.5 in language models. Now, it\u2019s leading the vision domain with OmniHuman and Goku excelling in 3D modeling and video synthesis. With Step-Video-T2V, China directly challenges top text-to-video models like Sora, Veo 2, and Movie Gen. Developed by Stepfun AI, [&hellip;]\" \/>\n<meta property=\"og:url\" content=\"https:\/\/fivemor.com\/?p=101583\" \/>\n<meta property=\"og:site_name\" content=\"Som2ny Network\" \/>\n<meta property=\"article:published_time\" content=\"2025-02-21T13:55:40+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/fivemor.com\/wp-content\/uploads\/2025\/02\/Step-Video.webp.webp\" \/>\n\t<meta property=\"og:image:width\" content=\"872\" \/>\n\t<meta property=\"og:image:height\" content=\"473\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/webp\" \/>\n<meta name=\"author\" content=\"admin\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"admin\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"8 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\/\/fivemor.com\/?p=101583#article\",\"isPartOf\":{\"@id\":\"https:\/\/fivemor.com\/?p=101583\"},\"author\":{\"name\":\"admin\",\"@id\":\"https:\/\/fivemor.com\/#\/schema\/person\/b85e3c3dc0e1daea076524dc8810c371\"},\"headline\":\"China\u2019s New AI Video Star: Step-Video-T2V\",\"datePublished\":\"2025-02-21T13:55:40+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/fivemor.com\/?p=101583\"},\"wordCount\":1569,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\/\/fivemor.com\/#organization\"},\"image\":{\"@id\":\"https:\/\/fivemor.com\/?p=101583#primaryimage\"},\"thumbnailUrl\":\"https:\/\/fivemor.com\/wp-content\/uploads\/2025\/02\/Step-Video.webp.webp\",\"keywords\":[\"Chinas\",\"Star\",\"StepVideoT2V\",\"Video\"],\"articleSection\":[\"Analytics\"],\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\/\/fivemor.com\/?p=101583#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/fivemor.com\/?p=101583\",\"url\":\"https:\/\/fivemor.com\/?p=101583\",\"name\":\"China\u2019s New AI Video Star: Step-Video-T2V - Som2ny Network\",\"isPartOf\":{\"@id\":\"https:\/\/fivemor.com\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/fivemor.com\/?p=101583#primaryimage\"},\"image\":{\"@id\":\"https:\/\/fivemor.com\/?p=101583#primaryimage\"},\"thumbnailUrl\":\"https:\/\/fivemor.com\/wp-content\/uploads\/2025\/02\/Step-Video.webp.webp\",\"datePublished\":\"2025-02-21T13:55:40+00:00\",\"breadcrumb\":{\"@id\":\"https:\/\/fivemor.com\/?p=101583#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/fivemor.com\/?p=101583\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/fivemor.com\/?p=101583#primaryimage\",\"url\":\"https:\/\/fivemor.com\/wp-content\/uploads\/2025\/02\/Step-Video.webp.webp\",\"contentUrl\":\"https:\/\/fivemor.com\/wp-content\/uploads\/2025\/02\/Step-Video.webp.webp\",\"width\":872,\"height\":473},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/fivemor.com\/?p=101583#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\/\/fivemor.com\/?bp_activities=1\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"China\u2019s New AI Video Star: Step-Video-T2V\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/fivemor.com\/#website\",\"url\":\"https:\/\/fivemor.com\/\",\"name\":\"Som2ny Network\",\"description\":\"Daily Deals\",\"publisher\":{\"@id\":\"https:\/\/fivemor.com\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/fivemor.com\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Organization\",\"@id\":\"https:\/\/fivemor.com\/#organization\",\"name\":\"Som2ny Network\",\"url\":\"https:\/\/fivemor.com\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/fivemor.com\/#\/schema\/logo\/image\/\",\"url\":\"https:\/\/fivemor.com\/wp-content\/uploads\/2026\/07\/4a0953c4-logo-300x86-1.png\",\"contentUrl\":\"https:\/\/fivemor.com\/wp-content\/uploads\/2026\/07\/4a0953c4-logo-300x86-1.png\",\"width\":300,\"height\":86,\"caption\":\"Som2ny Network\"},\"image\":{\"@id\":\"https:\/\/fivemor.com\/#\/schema\/logo\/image\/\"}},{\"@type\":\"Person\",\"@id\":\"https:\/\/fivemor.com\/#\/schema\/person\/b85e3c3dc0e1daea076524dc8810c371\",\"name\":\"admin\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/fivemor.com\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/secure.gravatar.com\/avatar\/729ae85bf62b9917e93538db2f2688ca?s=96&r=g&default=https%3A%2F%2Ffivemor.com%2Fwp-content%2Fplugins%2Fbuddypress-first-letter-avatar%2Fimages%2Fdefault%2F96%2Flatin_a.png\",\"contentUrl\":\"https:\/\/secure.gravatar.com\/avatar\/729ae85bf62b9917e93538db2f2688ca?s=96&r=g&default=https%3A%2F%2Ffivemor.com%2Fwp-content%2Fplugins%2Fbuddypress-first-letter-avatar%2Fimages%2Fdefault%2F96%2Flatin_a.png\",\"caption\":\"admin\"},\"sameAs\":[\"https:\/\/fivemor.com\"],\"url\":\"https:\/\/fivemor.com\/?author=1\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"China\u2019s New AI Video Star: Step-Video-T2V - Som2ny Network","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/fivemor.com\/?p=101583","og_locale":"en_US","og_type":"article","og_title":"China\u2019s New AI Video Star: Step-Video-T2V - Som2ny Network","og_description":"China is advancing rapidly in generative AI, building on successes like DeepSeek models and Kimi k1.5 in language models. Now, it\u2019s leading the vision domain with OmniHuman and Goku excelling in 3D modeling and video synthesis. With Step-Video-T2V, China directly challenges top text-to-video models like Sora, Veo 2, and Movie Gen. Developed by Stepfun AI, [&hellip;]","og_url":"https:\/\/fivemor.com\/?p=101583","og_site_name":"Som2ny Network","article_published_time":"2025-02-21T13:55:40+00:00","og_image":[{"width":872,"height":473,"url":"https:\/\/fivemor.com\/wp-content\/uploads\/2025\/02\/Step-Video.webp.webp","type":"image\/webp"}],"author":"admin","twitter_card":"summary_large_image","twitter_misc":{"Written by":"admin","Est. reading time":"8 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/fivemor.com\/?p=101583#article","isPartOf":{"@id":"https:\/\/fivemor.com\/?p=101583"},"author":{"name":"admin","@id":"https:\/\/fivemor.com\/#\/schema\/person\/b85e3c3dc0e1daea076524dc8810c371"},"headline":"China\u2019s New AI Video Star: Step-Video-T2V","datePublished":"2025-02-21T13:55:40+00:00","mainEntityOfPage":{"@id":"https:\/\/fivemor.com\/?p=101583"},"wordCount":1569,"commentCount":0,"publisher":{"@id":"https:\/\/fivemor.com\/#organization"},"image":{"@id":"https:\/\/fivemor.com\/?p=101583#primaryimage"},"thumbnailUrl":"https:\/\/fivemor.com\/wp-content\/uploads\/2025\/02\/Step-Video.webp.webp","keywords":["Chinas","Star","StepVideoT2V","Video"],"articleSection":["Analytics"],"inLanguage":"en-US","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/fivemor.com\/?p=101583#respond"]}]},{"@type":"WebPage","@id":"https:\/\/fivemor.com\/?p=101583","url":"https:\/\/fivemor.com\/?p=101583","name":"China\u2019s New AI Video Star: Step-Video-T2V - Som2ny Network","isPartOf":{"@id":"https:\/\/fivemor.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/fivemor.com\/?p=101583#primaryimage"},"image":{"@id":"https:\/\/fivemor.com\/?p=101583#primaryimage"},"thumbnailUrl":"https:\/\/fivemor.com\/wp-content\/uploads\/2025\/02\/Step-Video.webp.webp","datePublished":"2025-02-21T13:55:40+00:00","breadcrumb":{"@id":"https:\/\/fivemor.com\/?p=101583#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/fivemor.com\/?p=101583"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/fivemor.com\/?p=101583#primaryimage","url":"https:\/\/fivemor.com\/wp-content\/uploads\/2025\/02\/Step-Video.webp.webp","contentUrl":"https:\/\/fivemor.com\/wp-content\/uploads\/2025\/02\/Step-Video.webp.webp","width":872,"height":473},{"@type":"BreadcrumbList","@id":"https:\/\/fivemor.com\/?p=101583#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/fivemor.com\/?bp_activities=1"},{"@type":"ListItem","position":2,"name":"China\u2019s New AI Video Star: Step-Video-T2V"}]},{"@type":"WebSite","@id":"https:\/\/fivemor.com\/#website","url":"https:\/\/fivemor.com\/","name":"Som2ny Network","description":"Daily Deals","publisher":{"@id":"https:\/\/fivemor.com\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/fivemor.com\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Organization","@id":"https:\/\/fivemor.com\/#organization","name":"Som2ny Network","url":"https:\/\/fivemor.com\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/fivemor.com\/#\/schema\/logo\/image\/","url":"https:\/\/fivemor.com\/wp-content\/uploads\/2026\/07\/4a0953c4-logo-300x86-1.png","contentUrl":"https:\/\/fivemor.com\/wp-content\/uploads\/2026\/07\/4a0953c4-logo-300x86-1.png","width":300,"height":86,"caption":"Som2ny Network"},"image":{"@id":"https:\/\/fivemor.com\/#\/schema\/logo\/image\/"}},{"@type":"Person","@id":"https:\/\/fivemor.com\/#\/schema\/person\/b85e3c3dc0e1daea076524dc8810c371","name":"admin","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/fivemor.com\/#\/schema\/person\/image\/","url":"https:\/\/secure.gravatar.com\/avatar\/729ae85bf62b9917e93538db2f2688ca?s=96&r=g&default=https%3A%2F%2Ffivemor.com%2Fwp-content%2Fplugins%2Fbuddypress-first-letter-avatar%2Fimages%2Fdefault%2F96%2Flatin_a.png","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/729ae85bf62b9917e93538db2f2688ca?s=96&r=g&default=https%3A%2F%2Ffivemor.com%2Fwp-content%2Fplugins%2Fbuddypress-first-letter-avatar%2Fimages%2Fdefault%2F96%2Flatin_a.png","caption":"admin"},"sameAs":["https:\/\/fivemor.com"],"url":"https:\/\/fivemor.com\/?author=1"}]}},"_links":{"self":[{"href":"https:\/\/fivemor.com\/index.php?rest_route=\/wp\/v2\/posts\/101583","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/fivemor.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/fivemor.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/fivemor.com\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/fivemor.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=101583"}],"version-history":[{"count":0,"href":"https:\/\/fivemor.com\/index.php?rest_route=\/wp\/v2\/posts\/101583\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/fivemor.com\/index.php?rest_route=\/wp\/v2\/media\/101584"}],"wp:attachment":[{"href":"https:\/\/fivemor.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=101583"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/fivemor.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=101583"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/fivemor.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=101583"},{"taxonomy":"dealstore","embeddable":true,"href":"https:\/\/fivemor.com\/index.php?rest_route=%2Fwp%2Fv2%2Fdealstore&post=101583"},{"taxonomy":"offerexpiration","embeddable":true,"href":"https:\/\/fivemor.com\/index.php?rest_route=%2Fwp%2Fv2%2Fofferexpiration&post=101583"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}