Browser Use 开源的会话式视频剪辑技能包 video-use（当日涨星 +745、★26,450、MIT）：不让模型看像素，只让它读 12KB 词级转写文本 + 按需抽查画面，配合 12 条硬规则与 EDL 驱动的 ffmpeg 确定性渲染管线。拆解两条读入通道、六步流水线与自评回路，附五个落地场景与成本口径。

Internal emails and sworn testimony undercut OpenAI and Microsoft's fair use defense. A Microsoft director described the…

Computer-use agents (CUAs) have advanced along two separate lines: graphical interaction and software development throug…

Computer-use agents are increasingly evaluated in realistic desktop environments, but existing benchmarks provide limite…

Computer-use agents increasingly interact with browsers, terminals, file systems, and external services, introducing saf…

Alibaba's Qwen3.8-Omni-Flash understands audio and video, plans tasks, calls tools, and reports about 45.7% fewer tokens…

People across the world are increasingly turning to AI for everyday tasks, from looking up recipes and doing their homew…

YC 背景团队开源的计算机使用代理基础设施 Cua（当日涨星 1,112、★24,318、MIT）：Driver 提供带 7 类 oracle 证据链的 GUI 工具面，Sandbox/Fleet 提供可复现的隔离电脑，Cua-Bench 提供评测与轨迹，2.8 MB 的 CUA-S1 打分器替代部分大模型调用。含五个落地场景与可复现命令。

If AI lab PrismML isn't on your radar yet, it should be.]

At its Made On YouTube annual event, YouTube announced new tools for YouTube Studio, the app creators use to manage thei…

In a bid to help creators reach more people, YouTube on Wednesday unveiled a slew of new tools, many of which use AI to …

But the military's overall use of AI seems to be accelerating.]

A walking robotic hand must use the same fingers to move its body, support its weight, and interact with the environment…

Large language model (LLM) agents allocate test-time compute adaptively as they revise solutions, use tools, explore alt…

AI inference chipmaker d-Matrix today announced it will use NVLink Fusion to connect its next-generation Raptor XPUs to …

Co-work agents execute complex workflows that combine information gathering, tool use, coding, and file manipulation acr…

We investigate what makes synthetic OCR supervision transfer to real Thai documents and use the resulting insights to bu…

Multi-agent LLM systems commonly use an orchestrator to decompose a task for a team of workers and then improve through …

OpenAI announced on Wednesday that it is bringing voice-based agentic features to mobile, allowing users to trigger work…

NVIDIA has released Nemotron 3 Diarization, an open-weight speaker diarization model on Hugging Face. It answers one que…

Nokia’s applied research team has open-sourced AnyJev, a Python library that turns an open LLM into a decision model. It…

Sep 23, 2026 Meta Key Points Meta's new AI agent Muse drew more than 500,000 users in its first week in the US. The app …

Alibaba's Qwen team has released Qwen-Image-2.1, a 7B diffusion transformer that handles text-to-image generation, multi…

SpaceXAI has released Grok 4.7, its new flagship model for coding, agentic tasks, and knowledge work. Grok 4.7 is built …

Google’s AI-native Googlebook ties Gemini to the cursor, dictation, widgets, and other parts of the desktop experience.]

Flet 1.0 shipped on September 15, 2026, and the team now calls the framework ready for production apps. We look at what …

TypeSafe AI 的 Jev 全面开放，注册即得 5 美元额度（约 1.2 亿输入 Token），输出 Token 永久免费。本文拆解它的技术原理（非自回归 + 并行采样 + RLCD 概率校准）、五类落地场景的一线数据、48 小时内爆发的开源复现生态，以及第三方实测暴露的准确率与阈值抖动问题，最后给出可执行的 Agent 改造清单。

OpenClaw 2026.9.5 ships 4,179 pull requests from 502 contributing accounts. The headline change is Atomic Updates, which…

9月19日，中国人工智能企业APUS旗下 AI 实验室公布了全球最早一批针对Jev的独立开源复现成果]

TypeSafe AI released Jev, a System One model that answers typed questions with probabilities instead of generating text.…

PrismML has released Ternary Bonsai 2 27B, a ternary-weight version of Qwen3.8 27B. The language model occupies 5.93 GB,…

Petlibro's new Granary 2 smart feeders use a built-in scale and (on pricier models) an AI camera to track exactly how mu…

We present MintAct, a family of vision-language models that unifies UI grounding, multi-step navigation across mobile, d…

Paper2Agent, published in Nature, converts papers into validated MCP tools, scoring 91.2% on 300 questions across 74 pap…

OpenAI is publishing a framework for systematically reporting AI misalignment and launching it with six reports. In one …

GLiFormer Large scores 91.10 F1 on nested JSON, near GPT-5.6-luna's 91.96, while grounding every value in source spans. …

Ursula von der Leyen plans to invite the major frontier labs to talks and use the AI Act to help set global AI safety st…

Nums AI has released Causilo, a pretrained tabular foundation model for classification and regression with a scikit-lear…

Learn how to leverage NVIDIA’s cuDNN Frontend Graph API to build custom kernel fusions, autotuning engine configurations…

· Google 向全体工程师开放 Claude，Gemini 仍是默认模型 · 鸿蒙智行确认问界合作模式调整，赛力斯主导五项业务 · Waymo 计划 2027 年在东京推出全无人出租车 #欢迎关注爱范儿官方微信公众号：爱范儿（微信号：i…

GPT-6 + Hyper3D MCP 最强3D创作搭档 #欢迎关注爱范儿官方微信公众号：爱范儿（微信号：ifanr），更多精彩内容第一时间为您奉上。 ]

Sep 23, 2026 Nano Banana Pro prompted by THE DECODER Key Points Google has released two new text-to-speech models, Gemin…

Google has released Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS, 2 new text-to-speech models in its Gemini Audio …

Sep 23, 2026 YouTube is rolling out new AI tools for creators.** A storytelling assistant in YouTube Studio analyzes scr…

Confirming earlier reports, Meta announced its first pair of camera-free AI glasses, the clunkily named Ray-Ban Meta Aud…

At a glance Challenges a core assumption in robotics AI: Our research shows that running physical AI inference exclusive…

Sep 23, 2026 ChatGPT Voice gets a major upgrade.** The globally available voice feature now runs on OpenAI's new GPT-6 A…

Early adopters of Meta’s Muse have been speculating that the reason the AI works so well is because it’s OpenClaw under …

克雷西* 2026-09-23 15:29:50 来源：量子位 每秒能产生5000+个沙盒 克雷西 发自 凹非寺 量子位 | 公众号QbitAI 大模型训练拼的是算力，Agent训练拼的是环境。 环境怎么造？梁文锋署名的DeepSeek最…

NVIDIA AI Day Singapore, which takes place Sept. 22-23 at the Raffles City Convention Centre, is offering attendees oppo…