Workflow
MirageLSD
icon
Search documents
AI与机器人盘前速递丨优必选拿下人形机器人企业最大采购订单;全球首个直播生成模型发布
Mei Ri Jing Ji Xin Wen· 2025-07-21 01:23
3、上海人工智能实验室于7月19日发布DeepLink超大规模跨域混训技术方案,并已完成多个项目落地, 支持千公里多智算中心跨域长稳混训千亿参数大模型,标志着超大规模智算跨省互联实现新突破。 【市场复盘】 上周五(7月18日),截至收盘,科创人工智能ETF华夏(589010)收涨0.39%,持仓股福昕软件领涨 4.99%,合合信息上涨4.24%,优刻得、金山办公涨幅超3%;机器人ETF(562500)收平盘,盘中呈震 荡走势,多空力量较为均衡。中大力德领涨5.04%,东杰智能上涨3.74%,巨轮智能、科瑞技术涨幅超 2%;瀚川智能、石头科技、晶品特装下跌超2%。当日交易金额8.96亿元,换手5.75%,回切稳健风格, 市场坚定持有静待后续行情。规模方面,上周机器人ETF规模迭创新高,增加7.30亿元,最新规模达 155.29亿元,超过市场同类基金的规模总和,"吸金"能力出类拔萃;机器人ETF最新份额达177.15亿 份,位居可比基金首位。 【热点要闻】 1、近日,优必选科技中标觅亿(上海)汽车科技有限公司价值9051.15万元的机器人设备采购项目,创 下全球人形机器人企业最大中标金额纪录。 此前一天,优必选发 ...
腾讯研究院AI速递 20250721
腾讯研究院· 2025-07-20 16:02
生成式AI 一、 几千人盲投Kimi K2超越DeepSeek拿下全球开源第一 1. Kimi K2在最新排名中超越DeepSeek成为全球开源模型第一,总榜排名第五,紧追顶尖闭 源模型; 2. K2继承了DeepSeek V3架构并进行参数调整,包括增加专家数量、减半注意力头数、保 留第一层Dense及专家无分组; 3. 全球TOP 10开源模型中唯二入选的均来自中国,"开源=性能弱"的印象正被打破。 https://mp.weixin.qq.com/s/rTa_dKgKx40zRXJtHkeqgA 二、 世界首个「实时、无限」扩散视频模型,Karpathy投资 1. Decart发布MirageLSD,首个实时(40毫秒延迟)、无时长限制的扩散视频模型,可处理任 意视频流; 2. Karpathy成为天使投资人,预见其在实时电影制作、游戏开发和AR领域的广泛应用; 3. 技术突破在于实时流扩散(LSD)架构,通过逐帧生成和历史增强方法解决误差累积问题,但 精细控制和几何稳定性仍需改进。 https://mp.weixin.qq.com/s/yeWZCjtEBXmJaHsa8mf54w 三、 Suno V4 ...
大神Karpathy都投的AI实时视频生成模型:直播都能立即转,无限时长几乎零延迟
量子位· 2025-07-19 05:15
Core Viewpoint - The article discusses the innovative AI startup Decart and its groundbreaking video model MirageLSD, which enables real-time, zero-latency video generation, revolutionizing live streaming, gaming, and video communication [4][5][7]. Group 1: Technology and Features - MirageLSD is the first AI model to achieve zero-latency, infinite real-time video generation, allowing for continuous video streams without time limitations [4][5]. - The model operates at a speed 16 times faster than previous models, generating video at 24 frames per second and allowing for ongoing prompts, transitions, and edits during video generation [6][28]. - It addresses the "error accumulation" issue found in traditional autoregressive video models, ensuring temporal coherence while generating content frame by frame [9][11]. Group 2: Innovations and Mechanisms - The model employs a custom real-time stream diffusion model (Live-Stream Diffusion) that generates each frame based on previously generated frames and user prompts, rather than relying on the entire video sequence [14]. - It utilizes Diffusion Forcing technology to independently denoise single frames during training, ensuring coherence in frame generation [15]. - The model incorporates a historical enhancement strategy to preemptively correct potential errors by simulating artifacts during training [16]. Group 3: Performance and User Interaction - MirageLSD's architecture includes an improved Transformer model and a specially designed visual encoder, which enhances processing speed and reduces latency [18][20]. - The system features a dynamic input mechanism that processes player inputs with ultra-low latency, allowing for immediate responses to changes in the environment [22]. - Users can perform actions like changing outfits or transforming objects with minimal delay, showcasing the model's interactive capabilities [23]. Group 4: Company Background and Future Developments - Decart, the company behind MirageLSD, was founded in 2023 and previously launched the Oasis model, which also supports real-time interactions [25][26]. - The team plans to regularly release upgrades and new features for MirageLSD, including facial consistency, voice control, and precise object manipulation to enhance user experience [28].
世界首个「实时、无限」扩散视频生成模型,Karpathy投资站台
机器之心· 2025-07-19 03:13
机器之心报道 但如果加上两个关键词,这将成为 AI 视频生成领域革命性的突破! 就在昨天,Decart 发布了世界上首个 「实时的」「无时长限制的」 并且支持「任意视频流」的扩散视频模型 MirageLSD! 输入任何视频流,无论是相机或视频聊天、电脑屏幕还是游戏,MirageLSD 都能在 40 毫秒延迟 以内 将其转化为你想要的任何世界。 这一切都看上去不可思议,AI 视频已经能够实现和滤镜一样的应用方式,实时智能调整画面风格和画面内容,并且能够通过文本提示任意地进行控制。 实时视频魔法 解锁全新应用可能 前特斯拉 AI 总监,OpenAI 的创始团队成员 Andrej Karpathy 为此技术展开了广泛的想象: 编辑:冷猫 一觉起来世界已经进化成这样了? 每个人都能懂点魔法,能够随意穿梭在各个平行时空和幻想世界里。 读者朋友们看到这说不定撇撇嘴,「这不就是 AI 视频吗?」 1. 将 摄像头画面 变为 "另一个世界"。 2. 自导自演 实时电影 :拿起道具、演绎场景,AI 负责实时布景和风格化,秒看回放,边演边剪。 3. 游戏开发 轻松起步:用简单的球体 / 方块编码游戏机制,再用实时扩散模型为游戏生 ...