June 18, 2025 05:12 AM GMT Kuaishou Technology | Asia Pacific AI Video Generation Competition Increases with a Strong Debut of Seedance 1.0 Pro We see a change in the competition of AI video generation with the release of two new models recently. What is new: On 11 June, ByteDance released its AI video generation model Seedance 1.0 pro in its Volcano Engine Force Conference. Seedance 1.0 now ranks No.1 on the Artificial Analysis Leaderboard for both text-to-video and image-to-video, surpassing Google's Veo ...

FVDM & Pusa 一作：刘耀芳目前在香港城市大学攻读博士学位，导师为著名数学家 Prof. Raymond Chan (陈汉夫) 及 Prof. MOREL Jean-Michel。他 也曾在腾讯 AI Lab 实习，主导 / 参与 EvalCrafter , VideoCrafter 等工作，其研究兴趣包括扩散模型，视频生成等；项目主管：刘睿，香港中文大学 MMLab 博士，华为香港研究所小艺团队技术负责人。 论文标题：Redefining Temporal Modeling in Video Diffusion: The Vectorized Timestep Approach FVDM 论文：https://arxiv.org/abs/2410.03160 Pusa 主页 / 代码库: https://github.com/Yaofang-Liu/Pusa-VidGen 具体而言，Pusa 通过对预训练模型如 Wan-T2V 14B 进行 非破坏性微调，仅需 500 美金训练成本即可达到比 Wan 官方 I2V（至少 O(100k) 美金的训练 成本）更好的效果，成本降低超 200 倍，数据更是减 ...

This talk will briefly trace the history of video generation models before diving into Veo 3, Google DeepMind's latest state-of-the-art model that marks a significant leap by generating video with synchronized audio—including dialogue, sound effects, and music—all from text and image prompts. We'll show how it can understanding intricate details, maintain coherence over longer sequences, and simulate realistic physics and camera movements. For developers, Veo 3, accessible via Vertex AI (preview), unlocks m ...

整理 | 褚杏娟 近日，字节分享了其对今年 AI 技术发展主线的思考，主要包括以下三个方面： 基于上述考虑，6 月 11 日，字节跳动旗下火山引擎进行一系列的发布和更新，包括豆包大模型 1.6、视频生成模型 Seedance 1.0 pro 等新模型，并升级了 Agent 开发平台等 AI 云原生服务。 豆包 1.6 实行统一定价 大会上，字节发布了豆包大模型 1.6，包括 Doubao-Seed-1.6-thinking、Doubao-Seed-1.6、 Doubao-Seed-1.6-flash，均支持多模态输入，并实现 256K 超长上下文。 Doubao-Seed-1.6 支持 auto/thinking/non-thinking 三种思考模式。据介绍，在高考全国新一卷数学 中，豆包大模型取得 144 分；在海淀模拟全卷考试中，理科取得 706 分，文科获得 712 分。 豆包 1.6 系列模型支持多模态理解和图形界面操作，能够理解和处理真实世界问题。演示案例显示， 豆包 1.6 可自动操作浏览器完成酒店预定，识别购物小票并整理成 Excel 表格等任务。 | 候型能力 | | | | | | | - ...

1. 【格力电器：公司芯片已在家用空调产品中规模应用，整体自研应用占比约30%】格力电器董事 会秘书章周虎在6月9日下午举行的公司业绩说明会上表示，目前公司芯片产品主要有功率半导体产品 以及集成电路芯片，"公司芯片已大规模应用在家用空调产品中，整体自研应用占比约30%，自研芯 片也已应用在商用空调、智能装备、工业机器人等自营业务中。"注：格力自2015年开始进入芯片领 域，一期规划产能24万片/年。（财联社） O AIGC 产业动态日报 谷歌宣布在医疗保健领域推出人工智能计划 并致力于打造/ Stability Al发布3D视频生成工具SV3D! 可 同时输出多个新视角 这款新模型基于Stable Video Diff on模型的改讲,留 移根据单一辑入图像创建和转换多视图3D网格。 2.【 台积电日本、德国新厂建设计划或将调整 】6月9日消息，台积电董事长魏哲家近日表示，台积 电日本熊本二厂建设确实延宕，但主因是"塞车问题"。供应链透露，在资金、人力排挤下，加上评估 当地客户下单状况，目前台积电日本熊本一厂产能利用率尚未达标，而德国厂兴建与后续进机计划， 预估也将有所调整。整体来说，日本、欧洲现阶段车市不佳 ...

Created with Veo 3 - our new state-of-the-art video generation model, designed to empower filmmakers and storytellers. Veo 3 lets you add sound effects, ambient noise, and even dialogue to your creations – generating all audio natively. It also delivers best in class quality, excelling in physics, realism and prompt adherence. Find out more: https://deepmind.google/models/veo/ Prompt: The scene explodes with the raw, visceral, and unpredictable energy of a hardcore off-road rally, captured with a dynamic, a ...

Created with Veo 3 - our new state-of-the-art video generation model, designed to empower filmmakers and storytellers. Veo 3 lets you add sound effects, ambient noise, and even dialogue to your creations – generating all audio natively. It also delivers best in class quality, excelling in physics, realism and prompt adherence. Find out more: https://deepmind.google/models/veo/ Prompt: The scene opens with a top-down or wide-angle shot showcasing a vast, perfectly flat, neutral-colored surface – perhaps the ...

Created with Veo 3 - our new state-of-the-art video generation model, designed to empower filmmakers and storytellers. Veo 3 lets you add sound effects, ambient noise, and even dialogue to your creations – generating all audio natively. It also delivers best in class quality, excelling in physics, realism and prompt adherence. Find out more: https://deepmind.google/models/veo/ Prompt: In rural Ireland, circa 1860s, two women, their long, modest dresses of homespun fabric whipping gently in the strong coasta ...

Created with Veo 3 - our new state-of-the-art video generation model, designed to empower filmmakers and storytellers. Veo 3 lets you add sound effects, ambient noise, and even dialogue to your creations – generating all audio natively. It also delivers best in class quality, excelling in physics, realism and prompt adherence. Find out more: https://deepmind.google/models/veo/ Prompt: xxx ...

Created with Veo 3 - our new state-of-the-art video generation model, designed to empower filmmakers and storytellers. Veo 3 lets you add sound effects, ambient noise, and even dialogue to your creations – generating all audio natively. It also delivers best in class quality, excelling in physics, realism and prompt adherence. Find out more: https://deepmind.google/models/veo/ Prompt: A snow-covered plain of iridescent moon-dust under twilight skies. Thirty-foot crystalline flowers bloom, refracting light i ...