模型发布/更新
-
Google DeepMind Releases Gemini 3.8 Flash TTS and Flash-Lite TTS Speech Generation Models
Google DeepMind released two text-to-speech models, Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS, supporting…
-
Qwen Releases Qwen-Audio-3.1 Suite, Upgrades ASR, TTS and Realtime, Adds TTS-Next and ASR-Next
Qwen released Qwen-Audio-3.1, with comprehensive upgrades to ASR, TTS and Realtime, plus new audio creation model…
-
Fireworks Releases Ember-1, Matching Kimi K3 Quality with About 40% Fewer Tokens
Fireworks Research released Ember-1, a specialized model trained on Kimi K3 that learns to streamline unnecessary…
-
Arena Results: GPT-6 Sol (Max) Ranks 4th on Code Arena: WebDev with 1689 Points
Arena published real voting results for OpenAI's GPT-6 Sol (Max), which ranks 4th on the Code Arena: WebDev…
产品发布/更新
-
Greg Brockman 宣布 GPT Voice 大幅升级,支持使用工具并登陆 ChatGPT Work
GPT Voice 获得重大升级,现在可以使用邮箱、日历、Slack 等工具,并由 GPT-6 Astra、Sol 和 Luna 驱动。语音功能现已登陆网页端和移动端的 ChatGPT Work,用户可仅通过语音在浏览器中创建文档、演示文…
-
Antigravity SDK Supports Local Models, Debuting Gemma 4 26B A4B with LiteRT
Google announced that the Antigravity SDK supports local workflows, debuting support for running Gemma 4 26B A4B via…
-
OpenAI Opens Daybreak Cyber Defense Program to Ukrainian Government
OpenAI announced it is opening its Daybreak program to the Ukrainian government to support cyber defense of civilian…
-
Anthropic Launches Claude Marketplace: One-Stop Plugins, Agents and Service Partners
Anthropic launched Claude Marketplace, bringing plugins and connectors, Claude-powered agents and products, and…
-
Cursor Releases Rollouts and Security Reviewer Development Bots
Cursor released two software development bots, Rollouts and Security Reviewer, to help teams ship secure, reliable…
论文研究
-
Anthropic Establishes Life Sciences Lab; Claude Autonomously Discovers New CRISPR-Like Enzyme System ART
Anthropic announced the establishment of a life sciences research group and laboratory; a Claude agent, with only…
-
OpenAI Releases Open Benchmark MentalHealthBench with 80+ Mental Health Experts
OpenAI released MentalHealthBench, an open benchmark evaluating AI performance in real mental health conversations,…
技巧与观点
-
Team Shares Prompts for Improving Agent Harness Token Efficiency
A public prompt is designed to optimize LLM agent harnesses, aiming to reduce price-weighted token cost per task…
-
Anthropic 团队详解如何用 Claude 在两周内把 claude.ai 提速约 3 倍
Anthropic 团队发文详解今年 8 月通过两周冲刺让 claude.ai 和桌面端核心体验提速约 3 倍的过程:聚焦覆盖 95% 用户活动的四条旅程,75 分位首屏可输入时间从 3.1 秒降到 0.55 秒,合并超过三千个变更且无一…
-
Tomer Tunguz: AI's Most Important Market Is the Mid-Market, Not Frontier Models
Tomer Tunguz argues that enterprise AI usage is concentrated in the mid-market of multi-step workflows that require…
-
MiMo-V2.6-Pro, Claude Opus 5.5 and GPT-6 Luna/Sol Releases Shift Intelligence Index and Cost Pareto Frontier
Artificial Analysis said this week's releases of MiMo-V2.6-Pro, Claude Opus 5.5, GPT-6 Luna and GPT-6 Sol added eleven…
-
OpenRouter releases 2026 guide to choosing the best embedding models, covering 37 catalog entries
OpenRouter verified its embedding model catalog on September 11, 2026, finding 37 entries, and sent batch requests to…
-
AI 公司负责人在联合国安理会简报会上警告 AI 可能危及全人类
联合国安理会举行 AI 简报会,Yoshua Bengio、Sam Altman、Dario Amodei 和 Hugging Face CEO Clement Delangue 相继发言,美联社报道主要 AI 公司负责人警告若无干预,A…
-
Anthropic Field Engineers Share Six-Step Preparation Method for AI-Driven Code Modernization Projects
In its Notes from the Field series, Anthropic shared experience managing large code modernization projects, saying…