2026-08-14
AI 快讯日报.

Claude Code 会话如何最大化 token 价值 ↗

Practical tips for how to run efficient sessions that get the most value from every token.

OpenAI and Anthropic in price war as Chinese AI rivals gain ground ↗

Privacy Center Currently, only residents from GDPR countries and certain US states can opt out of Tracking Technologies through our Consent Management Platform. Additional options regarding these technologies may be available on your device, browser, or through industry options like AdChoices. Please see our Privacy Policy for more information.

GPT-5.6 Sol 推出 Ultrafast 模式，推理速度提升 14 倍，由 Cerebras 提供支持 ↗

OpenAI is launching "Ultrafast," a new inference mode that delivers GPT-5.6 Sol at up to 750 output tokens per second, powered by Cerebras hardware from their $10 billion partnership. Together with "Standard" and "Fast," Ultrafast creates a three-tier pricing structure that turns inference speed into its own product.

DeepSeek 峰谷电价更新：V4-Pro 正式发布，API 引入峰谷定价 ↗

We’re launching DeepSeek-V4-Pro today! 🚀

三星健康 AI 模型分析可穿戴生物信号数据 ↗

Samsung Research America’s Digital Health Team has presented two AI foundation models designed to learn from wearable biosignals. The work centres on data captured by smartwatches, including heart activity, sleep, and physical activity.

谷歌 Chrome 浏览器新功能曝光：Gemini 帮你修改弱密码和重复密码 ↗

谷歌 Chrome Canary 版正测试 AI 新功能，让 Gemini 自动帮你修改弱密码或重复密码，相比以往仅提醒用户，操作更便捷。

Claude Code now runs daily maintenance on Anthropic's software with a 46 percent merge rate ↗

Anthropic is testing whether Claude Code can handle daily maintenance of the company's own apps, from crash fuzzing to dead-code removal. In a few weeks, the AI created 388 pull requests, and 46 percent were merged after human review. Claude Code inventor Boris Cherny sees this as "early signs of life that this might be possible."

腾讯混元资深研究员徐灿转岗微信 WeLM，微信 AI 或进入加速阶段 ↗

据 36 氪消息，腾讯混元资深研究员徐灿已转岗至微信 WeLM 团队，主要负责大模型研发。此前他曾担任后训练方向负责人，此次转岗发生在微信强化自研大模型与 Agent 能力的阶段。

Apple 与阿里巴巴合作训练中国定制 AI 模型 ↗

The rare US-China partnership comes as Apple prepares to roll out its on-device generative AI service in China.

智谱发布 GLM-5.3，称其为最强开源权重编程模型 ↗

Zhipu AI has released GLM-5.3, a model that, according to its own benchmarks, is the most powerful open-weights coding model, with a 50 percent improvement over its predecessor through post-training alone. Trained for cybersecurity, GLM-5.3 helped security teams find 2,436 vulnerabilities across 269 projects. The model weights are set to go open source in two weeks.

dots3-note Preview 开源：280B 参数轻量模型，主打长程智能体与多模态推理 ↗

Google AI 健康教练将接入雅培血糖数据 ↗

Checking the site connection security

荣耀方飞：Robot Phone 只是第一代探索，未来手机将从"应用的容器"走向"智能体的舞台" ↗

方飞表示，荣耀一直认为未来手机会从“应用的容器”走向“智能体的舞台”。荣耀 Robot Phone 只是第一代探索，未来将会继续把想象变成用户真正能够感知、使用和信任的产品。

四位博主给千问发来Offer，把重复琐事交给它处理 ↗

70 天超 10 款模型：CNBC 称美国限制反而催化中国 AI 创新，加速进入全球企业业务体系 ↗

首尔经济日报今天（8 月 14 日）发布博文，报道称自今年 6 月以来，9 家中国 AI 公司已推出 10 款 AI 模型，而且在推理、编程和智能体等关键领域正逐步逼近美国顶尖闭源模型。

Z.ai 发布 GLM-5.3：不重训基座模型，复杂编程与长程任务能力显著提升 ↗

Z.ai just released GLM-5.3. GLM-5.3 runs on the same 743B base model as GLM-5.2. Every reported gain comes from scaled post-training: more task environments, more environment types, longer training. The results land in two places. Coding jumps most on the longest-horizon benchmarks, with Terminal-Bench 3.0 moving from 4.6 to 28.3. Cybersecurity moved further than Z.ai says it expected, with CyberGym reaching 84.5%. Weights are not public yet.

组织如何使用人工智能：来自ChatGPT的证据 ↗

芯擎科技车规级7纳米AI加速芯片"天工100"量产，硬件成本降至千元级 ↗

芯擎科技自研的 7 纳米车规级 AI 加速芯片“天工 100”已全面量产，可插卡式升级，无需改动原车硬件，兼容多品牌车型，为端侧大模型提供国产算力底座。

OpenAI 首席营收官丹尼斯离职，Wiz 总裁达利·拉伊奇接任 ↗

本周早些时候，长期任职的高管布拉德 · 莱特卡普宣布将离职开展新项目；负责 OpenAI 大部分核心业务的菲吉 · 西莫在休病假后于上月卸任；前产品负责人凯文 · 韦尔则已于 4 月离开 OpenAI。

MiniMax-Music3 音乐模型发布：AI 生成最长 5 分钟歌曲 ↗

AI 公司 MiniMax 今天（8 月 14 日）推出音乐生成模型 MiniMax-Music3，可根据歌词和音乐描述生成最长 5 分钟的完整歌曲。