资讯列表
MOSS-TTSD Makes a Stunning Open Source Debut: A Million Hours of Training Creates a New King in AI Podcasts
MOSS-TTSD (Text to Spoken Dialogue), developed by the Tsinghua University Speech and Language Laboratory (Tencent AI Lab) in collaboration with Shanghai Chuangzhi College, Fudan University, and Musi Intelligent, has been officially open-sourced. This marks a major breakthrough in AI speech synthesis technology for dialogue scenarios.This speech dialogue generation model is based on the Qwen3-1.7B-base model and is trained further using approximately 1 million hours of single-speaker voice data and 400,000 hours of dialog voice data.
8/2/2025 4:36:02 PM
AI在线
Amazon CEO Wants to Embed Ads in Alexa + Conversations to Open a New Business Model!
At the latest earnings call, Amazon CEO Andy Jassy revealed that the company is exploring the possibility of embedding advertisements in its latest AI assistant, Alexa . He mentioned that as users engage in deeper conversations with Alexa , ads could help consumers discover new products and create a new revenue stream for Amazon.Alexa is Amazon's upgrade to traditional voice assistants, aiming to provide a more natural and human-like interaction experience.
8/2/2025 4:36:00 PM
AI在线
Singapore AI startup SixSense Raises $8.5 Million in Funding
In the semiconductor industry, quickly identifying and resolving defects in the production process has always been a challenging issue. Singapore-based deep-tech startup SixSense is targeting this challenge by launching an AI-driven platform designed to help chip manufacturers predict and detect potential chip defects in real-time. Recently, SixSense successfully completed an $8.5 million Series A funding round, bringing its total funding to $12 million.
8/2/2025 4:35:58 PM
AI在线
Singapore AI startup SixSense Raises $8.5 Million in Funding
In the semiconductor industry, quickly identifying and resolving defects in the production process has always been a challenging issue. Singapore-based deep-tech startup SixSense is targeting this challenge by launching an AI-driven platform designed to help chip manufacturers predict and detect potential chip defects in real-time. Recently, SixSense successfully completed an $8.5 million Series A funding round, bringing its total funding to $12 million.
8/2/2025 4:35:58 PM
AI在线
Cline Raises $32 Million in Funding and Launches an Enterprise AI Coding Platform
Recently, Cline announced a successful fundraising of $32 million, with the latest Series A round led by Emergence Capital, and other participants include Pace Capital, 1984Ventures, Essence VC, Cox Exponential, and several well-known developer investors such as Jared Friedman, a partner at Y Combinator, and Eric Simons, CEO of Bolt.new.
8/2/2025 4:35:56 PM
AI在线
Microsoft Research Reveals: 20 Occupations Less Affected by AI, Including Massage Therapists and Housekeepers
With the rapid development of artificial intelligence, many people are worried that their industries may be affected, even facing the risk of unemployment. However, a recent Microsoft study shows that there are some occupations that are relatively safe in the short term and less likely to be threatened by artificial intelligence.
8/2/2025 4:35:53 PM
AI在线
Creao AI Completes $10 Million in Funding to Develop the Future Agent Collaboration Operating System
【AIbase Report】On July 31, 2025, in the United States, the Silicon Valley-based artificial intelligence startup Creao AI announced that it has successfully completed two rounds of financing: a multi-million-dollar seed round led exclusively by Yunqi, and a multi-million-dollar Pre-A round led by Monolith, with global top venture capital firms such as GL Ventures, HongShan, Yunqi, and Hua Capital participating.
8/2/2025 4:35:53 PM
AI在线
Kimi K2 High-Speed Version kimi-k2-turbo-preview Released, Outputting 40 Tokens per Second
Kimi-k2-turbo-preview, the high-speed version of Kimi K2, has been officially released.
8/2/2025 4:35:53 PM
AI在线
Kimi K2 High-Speed Version Released, Output Speed Increased to 40 Tokens per Second
The Kimi Open Platform has launched Kimi K2 Speed Edition. The new model is named kimi-k2-turbo-preview, with the same parameter scale as the existing kimi-k2, but the output speed has increased from 10 Tokens per second to 40 Tokens per second, significantly improving usage efficiency. This upgrade aims to optimize user experience and meet application scenarios requiring higher real-time performance.
8/2/2025 4:35:53 PM
AI在线
Kimi K2 High-Speed Version Released, Output Speed Increased to 40 Tokens per Second
The Kimi Open Platform has launched Kimi K2 Speed Edition. The new model is named kimi-k2-turbo-preview, with the same parameter scale as the existing kimi-k2, but the output speed has increased from 10 Tokens per second to 40 Tokens per second, significantly improving usage efficiency. This upgrade aims to optimize user experience and meet application scenarios requiring higher real-time performance.
8/2/2025 4:35:53 PM
AI在线
刷新3D生成上限！一键生成精细到毛发的3D资产
在高质量3D生成需求日益增长的背景下，如何高效生成结构精良、几何精细的三维资产，已成为AIGC和数字内容创作领域的关键挑战。 尽管近年来的3D生成方法发展迅速，精度越来越高，但现有框架在兼顾效率和质量上依然面临瓶颈，尤其是在高分辨率建模中，计算的高复杂度严重制约了生成速度和应用落地。 为解决上述困境，南洋理工大学联合数美万物、西湖大学提出了Ultra3D：一种全新的3D生成框架。
8/2/2025 1:50:44 PM
不圆
仅用提示词工程摘下IMO金牌！清华校友强强联手新发现，学术界不靠砸钱也能比肩大厂
鹭羽 发自 凹非寺. 量子位 | 公众号 QbitAI无需谷歌“钞能力”，两位清华校友强强联合，直接让基础模型Gemini 2.5 Pro轻松达到IMO金牌水平。 只需一个小小的提示词改动……该发现来自两位清华校友杨林和黄溢辰，他们共同设计了一套自我迭代验证流程和提示词优化，就成功让Gemini 2.5 Pro完成了今年IMO题目的解答。
8/2/2025 1:45:48 PM
不圆
AI Coding如何重构开发，模型×IDE×Agent深度对话｜量子位AI沙龙
Coding可谓是最受关注的AI落地场景之一。 独立开发者在用Vibe Coding，以更简单的方式让创意成为现实，企业也将AI Coding引入到工作流中，提高效率。 从模型到产品，从插件到AI原生IDE，从补全代码到自主编程，可以看到AI Coding已经以不同方式与形态进入到了日常生活、工作中。
8/2/2025 12:32:10 PM
林樾
AI Coding如何重构开发，模型×IDE×Agent深度对话｜量子位AI沙龙
Coding可谓是最受关注的AI落地场景之一。 独立开发者在用Vibe Coding，以更简单的方式让创意成为现实，企业也将AI Coding引入到工作流中，提高效率。 从模型到产品，从插件到AI原生IDE，从补全代码到自主编程，可以看到AI Coding已经以不同方式与形态进入到了日常生活、工作中。
8/2/2025 12:32:10 PM
林樾
AI答IMO难题坦承“不会”，OpenAI：这就是自我意识
时令 发自 凹非寺. 量子位 | 公众号 QbitAI这一次，“不会”竟成了大模型的高光时刻。 虽然在IMO第6题上得了零分，OpenAI的金牌模型却展现了“高智商的诚实”。
8/2/2025 12:13:04 PM
时令
小米浏览器接入豆包大模型，升级 “AI 搜索” 功能
AI在线 8 月 2 日消息，火山引擎昨日宣布，小米浏览器现已升级“AI 搜索”功能，接入豆包大模型及火山方舟高代码智能体产品。 同时，小米应用商店接入火山引擎扣子。 用户通过扣子搭建的智能体，可以上传到小米应用商店。
8/2/2025 7:53:36 AM
浩渺
多项力压 Grok 4、OpenAI o3，谷歌推出 Gemini 2.5 Deep Think 模型
AI在线 8 月 1 日消息，谷歌今晚宣布，Gemini 2.5 Deep Think 模型现已正式上线，现已在 Gemini 应用中向 Google AI Ultra 订阅用户开放。 这款新模型在多个关键基准测试中超越了 OpenAI 的 o3 和 xAI 的 Grok 4。 据外媒 Neowin 报道，今年 5 月，谷歌在 Google I/O 大会上首次介绍了 Gemini 2.5 Pro 的 Deep Think 模式，采用全新研究方法，可在作答前综合多个假设进行推理。
8/1/2025 10:18:20 PM
清源