Elon Musk@elonmuskAI 评分4848BREAKING: Grok 4.5 ranked #1 on LaurenBench with a score of 56.9%, ahead of Claude Sonnet 5, GLM 5.2, Claude Opus 5, Kimi K3 and GPT-5.6. The benchmark tests real-world AI agents across conversation, tool use, memory and safety.
SpaceXAI 昨日发布 Grok Voice Think Fast 2.0，是其最智能的语音对语音 AI 模型，每分钟音频费用 0.08 美元。该模型在语音到语音基准测试中质量指数达 82.9%，较 1.0 版提升 7.2 个百分点，可在说话时同步完成推理。
谷歌在 Google Flow Music 中推出 Lyria 3.5 音乐生成 AI 模型，重点提升旋律编排、歌词生成、人声表现及歌曲控制能力。新模型生成的人声更贴近真实演唱，情感层次更细腻，发音更准确，同时用户可更灵活地设定歌曲节奏和时长。
Rohan Paul@rohanpaul_aiAI 评分4747Today we’ve raised $52M Seed and we are announcing the public launch of S2.1 Pro. >It can clone a voice from 5 seconds of audio >2x faster than Cartesia & 1/6th the cost of Eleven Labs >most expressive model with word level control over emotion, intonation, pacing etc We support frontier AI companies including HeyGen, LiveKit, Retell, Sanas, and OpenArt all run our model in production. If you're a business and we can't cut your voice AI costs by 50%, we'll give you 1 year of Fish Audio for free. Book a demo: https://s.fish.audio/tmapke To celebrate our first birthday, we'll give you 1 month of S2.1 Pro for free. Like, retweet, and comment “Fish” to get it.
Google 在 Gemini API 更新日志中宣布 Gemini Robotics ER 2 进入公开预览，发布两个具身推理模型端点。
Thinking Machines Lab 发布开源权重模型 Inkling-Small，为 MoE 架构，总参数 276B、激活 12B，以约四分之一的体积达到与 Inkling 相当的性能。
推荐理由：官方给出了完整评测和效率曲线，读者可以据此评估这款小体积 MoE 模型在推理、编码和多模态任务上的性价比。
Microsoft AI 发布 MAI-Cyber-1-Flash，与 MDASH 搭配在 CyberGym 上取得 96%（any crash score），成本仅为现有最佳方案的 50%，可部署在 H100 上，负责 90% 的常规任务以便为最难 10% 保留 GPT 5.4。
推荐理由：官方公布了多款针对具体产品调优的 MAI Flash 模型及其在各业务线节省 GPU 成本的实测数字，可了解按任务分层选模型的做法。
Greg Brockman@gdbAI 评分4545After deployment, we applied GPT-5.6 Sol to advance the frontier of efficiency by making itself more efficient to run. The results: - 20% lower serving costs from production GPU kernel improvements. - 15%+ better token-generation efficiency from improved speculative decoding.
Elon Musk@elonmuskAI 评分5757SpaceXAI has released Grok Voice Think Fast 2.0 today, with the High reasoning variant debuting at #2 on the Artificial Analysis Speech to Speech Index at 82.9%, and #1 on Tau Voice for Agentic Performance at 56.5% - among the fastest models at 0.70s Time to First Audio Grok Voice Think Fast 2.0 is SpaceXAI's successor to Grok Voice Think Fast 1.0 (75.7% on the Speech to Speech Index). It is the only model in the Index's top five with an average Time to First Audio under 1 second, achieving 0.70 seconds vs. 1.14 seconds for the next fastest, GPT-Realtime-2 High. Key takeaways: ➤ Speech to Speech Index: Grok Voice Think Fast 2.0 High debuts at #2 at 82.9%, only behind Qwen Audio 3.0 Realtime Plus (84.1%) and ahead of GPT-Realtime-2.1 High (79.1%) and GPT-Realtime-2 High (77.2%). This is up 7.3 percentage points from Grok Voice Think Fast 1.0 (75.7%) ➤ Speech to Speech Index by Benchmark: On Tau Voice, Grok Voice Think Fast 2.0 High is the new leader at 56.5%, just ahead of Qwen Audio 3.0 Realtime Plus at 54.6% and its predecessor Grok Voice Think Fast 1.0 at 52.1%. On Big Bench Audio, it achieves 97.2%, behind leader Qwen Audio 3.0 Realtime Plus at 99.2%. On our Full Duplex Bench subset, it scores 95.1%, up from 77.8% for Grok Voice Think Fast 1.0, the largest driver of its index gain, behind Qwen Audio 3.0 Realtime Plus at 98.4% ➤ Speed: The model’s average Time to First Audio on Big Bench Audio is 0.70 seconds, faster than GPT-Realtime-2 High (1.14s), GPT-Realtime-2.1 High (1.21s), and Grok Voice Think Fast 1.0 (1.25s), and well ahead of Qwen Audio 3.0 Realtime Plus (4.02s) ➤ Price: Grok Voice Think Fast 2.0 is priced at $4.80 per hour of input audio, up from $3.00 for Grok Voice Think Fast 1.0 and more expensive than Qwen Audio 3.0 Realtime Plus ($4.42) and GPT-Realtime-2 High ($4.14), but ~2.2x cheaper than GPT-Realtime-2.1 High ($10.75) Congratulations @SpaceXAI @elonmusk! See below for more detail ⬇️
Google 发布 Lyria 3.5 音乐生成模型，可生成更自然的旋律、歌词和发音更清晰的人声。新功能“Selective Section Painting”允许用户编辑音轨的特定部分或将短旋律扩展为完整歌曲，无需从头开始。用户还可精确控制速度、时长（30 秒至 3 分钟）以及人声、鼓、贝斯等元素的时长。
Google DeepMind 今日在 Google Flow Music 中发布新一代音乐生成模型 Lyria 3.5，带来音乐性、歌词质量、人声表现力与创作控制的多项提升。新模型能生成更自然复杂的旋律结构，歌词对提示词的遵循度和结构意识更强，人声更逼真且富有情感，同时支持更便捷地控制输出节奏与时长。
推荐理由：Lyria 3.5 把音乐生成从玩具级推到了可用创作工具级别，歌词和情感人声的提升对内容创作者是实实在在的福音。
Poolside 发布 agentic 编码模型 Laguna S 2.1 并开源权重，5 月 22 日开始在 4,096 块 NVIDIA H200 上预训练，60 天后开放权重与全部评测轨迹。
推荐理由：文章梳理了 Laguna S 2.1 从预训练到开源六十天的过程与基准数据，并公开评测轨迹，读者可对比同规模模型的实际表现。
OpenAI 通过 API 推出 GPT Transcribe 和 GPT Live Transcribe 两款语音识别模型。GPT Transcribe 处理预录音频速度约为实时的 34 倍，词错误率为 3.31%，定价降至每分钟 $0.0045。
上海人工智能实验室 InternLM 发布基于 Mobius-v0 架构的 35B 基础模型 Intern-S2-Mobius，将知识组织为全局共享 Memory，由多个 Reasoner 迭代查询和精炼隐藏状态。
OpenAI 通过 API 推出 GPT Transcribe 和 GPT Live Transcribe 两款新语音识别模型。GPT Transcribe 处理预录音频的速度约为实时的 34 倍，词错误率为 3.31%，较前代降低 0.7 个百分点，定价降至每分钟 $0.0045。
Liquid AI 发布两款开源双向编码器 LFM2.5-Encoder-230M 和 350M，均支持 8,192 token 上下文窗口。350M 模型在 17 项任务上平均得分 81.02，超越 ModernBERT-base 和所有 EuroBERT 模型；在 CPU 上处理 8K token 时，230M 模型推理仅需约 28 秒。
Greg Brockman@gdbAI 评分44445.6 用于解决另一个长期存在的开放问题，这次是在概率论领域：
My friend Zhengqing used GPT-5.6 to solve a beautiful well-known conjecture in probability. Feige's 1/e conjecture: for independent nonnegative X_1,…,X_n with E[X_i] ≤ 1 and S_n = X_1+⋯+X_n, we have P(S_n ≤ E[S_n] + 1) ≥ 1/e. Exciting to see a problem we used to kick around over lunch and dinner get solved! 1/2
Artificial Analysis@ArtificialAnlysAI 评分6969
OpenAI Developers@OpenAIDevs精选AI 评分6565
推荐理由：OpenAI 正式把转录拆成实时和异步两个模型，这是一个信号，语音 AI 的落地正在从通用走向场景专用，做语音产品的都该看一眼。
Elon Musk@elonmuskAI 评分5151In our latest http://deepsec.sh benchmarks, Grok 4.5 has emerged as the best cybersecurity AI model on price-performance. It's 10x cheaper than Sol, 5.7x cheaper than Opus 5, and 2.2x cheaper than Kimi K3, yet at Kimi-like perf. Sol remains the frontier, ahead of Opus 5.
智元自研的具身原生全模态大模型 WITA-Omni Preview 以 85.21 分综合成绩登顶 DailyOmni 榜单，在八项细分指标中的六项取得第一，超过千问、Gemini、豆包、英伟达等模型。该模型采用 Thinker–Talker–Actor 架构，将物理动作和表情提升为与语音并列的一级输出，并通过千万小时级多模态数据训练实现音视频联合理解与时序推理。
Microsoft 发布 MAI-Cyber-1-Flash，一款 137B 总参数（5B 活跃参数）、256k 上下文窗口的稀疏 MoE 网络安全模型，是 MAI-Code-1-Flash 的微调版本。
推荐理由：微软这个安全模型把漏洞挖掘系统推到95.95%，关键是90%任务由5B模型完成成本砍半，路由设计比模型本身更值得关注，做漏洞挖掘的可以马上跑起来试试。
AMD 推出 Instella-MoE 系列开源模型，总参数量 16B、激活参数量 2.8B，共 6 个版本，基于 AMD Instinct GPU 和 ROCm 软件栈从头训练。预训练阶段通过 FarSkip-Collective 将速度提升 12.7%，推理阶段使用 SGLang 框架将首次 Token 响应时间最多缩短 39.2%。
月之暗面发布 2.8 万亿参数的混合专家模型 Kimi K3，支持原生视觉理解和 100 万 token 上下文窗口。其规模化效率较 Kimi K2.5 提升 2.5 倍，并同步开源模型权重、技术报告及 MoonEP、FlashKDA、AgentEnv 三项 Infra 技术。
推荐理由：Kimi K3 的完全开放是一个明确信号，月之暗面把最核心的模型权重和技术细节全部公开，国内开发者从此有了媲美国际顶尖开源的底座。虽然部署门槛不低，但开放本身推动了生态发展。
Artificial Analysis@ArtificialAnlysAI 评分7070微软发布首个网络安全 AI 模型 MAI-Cyber-1-Flash，在 CyberGym 基准测试中取得 95.95% 的成功率，超过 Claude Mythos 5 和 GPT-5.5 Cyber。该模型接入多智能体安全系统 MDASH，可处理约 90% 的网络安全任务，与 GPT-5.4 的组合成本比此前配置降低近 50%。
Rohan Paul@rohanpaul_aiAI 评分4747Today, we are announcing a series of updates that give customers frontier-grade security at half the cost. MAI-Cyber-1-Flash is our first cybersecurity model, built ground up to find the most challenging vulnerabilities in complex code bases. When combined with MDASH, it delivers world-class performance at 50 percent of the cost of leading models. We are bringing this capability to market through Project Perception, a complete agentic security offering grounded in real-world signals and security workflows. Teams of specialized agents work together to simulate attacks, detect and triage/investigate, and fix and remediate. This is the benefit of building the harness, context/signals, and action space separate from one model family. By combining specialized models and data with the right agents, tools, security context, and harness, we can advance the frontier of cost to outcome.
Moonshot AI 发布了 Kimi K3 的模型权重与技术报告，并在 Hugging Face 上开源了高性能注意力内核、MoE 通信库等基础设施组件。新架构声称每单位算力的智能度提升 2.5 倍。该模型在基准测试中接近 Fable 5 和 GPT-5.6 Sol 等西方前沿模型，但英国网络研究所的独立测试显示其网络能力和数学技能仍落后于前沿模型。
Rohan Paul@rohanpaul_aiAI 评分5757Releasing the model weights and technical report of Kimi K3. Kimi K3 is our most capable model: a 2.8T MoE model with native visual understanding and a 1M-token context window. New model architecture: 2.5x the intelligence per unit of compute, not just more params. Alongside Kimi K3, we're opening up more of the stack behind it — high-performance attention kernels, MoE communication library, and infrastructure for running agent environments at scale. Model weights: http://huggingface.co/moonshotai/Kimi-K3 Tech report: http://github.com/MoonshotAI/Kimi-K3/blob/master/k3_tech_report.pdf Tech blog: http://kimi.com/blog/kimi-k3
微软推出其首个网络安全专用模型 MAI-Cyber-1-Flash，声称在 Cyber Gym 基准测试中击败了 Gemini、GPT 5.5 Cyber 等竞品。该模型与 GPT 5.4 绑定在 MDASH 框架内，用于发现复杂代码库中的漏洞。同时发布的 Perception 平台可部署红、蓝、绿三组 AI 智能体，预览版将于 11 月 3 日上线。
微软推出紧凑型安全模型 MAI-Cyber-1-Flash，集成于此前公布的 MDASH 多智能体系统。该组合在 CyberGym 基准上得分 96%，领先 Mythos 12 个百分点，并超越 Gemini 和 GPT。MAI-Cyber-1-Flash 处理 90% 的任务，仅将复杂案例转交 GPT-5.4，微软称成本可降低 50%。
Kimi.ai@Kimi_MoonshotAI 评分4141Kimi K3 is now live on Together AI. We’re proud to be a Day 0 launch partner for @Kimi_Moonshot’s open frontier model, built for long-running agentic workflows across code, tools, vision, and research.
Mustafa Suleyman@mustafasuleymanAI 评分5656月之暗面正式开源 Kimi K3 模型，该模型拥有 2.8 万亿参数、100 万 token 上下文窗口，并原生支持视觉理解。Kimi K3 基于 KDA 混合线性注意力机制构建，结合 Stable LatentMoE 框架在 896 个专家中激活 16 个，整体扩展效率相比 K2 提升约 2.5 倍。
