探HOW CONTENT DESK

资讯与观察

AI 行业动态、工程实践、产品方法和社区故事。

AI&大模型

突发!DeepSeek新模型发布,超级性价比斩杀一切

新版本V4.1-Flash-1017,还开放了自研搜索API

量子位10/171 分钟阅读站内导读

LATEST STORIES

最新内容

共 1522 篇
AI&大模型
量子位10/17

Claude半小时解开密码学悬案,学界集体走偏20年精选导读

暂无摘要

1 分钟64#大模型#国内动态#资讯
AI&大模型
The Decoder10/17

Deepseek V4.1 Flash 1017 beats larger rivals in agentic tasks while once again massively cutting per-task costs精选导读

Deepseek has released V4.1-Flash-1017 as an open-weight model that posts better benchmarks, finishes tasks faster, and produces shorter outputs than its predecessor. The company is also opening up its internal search en…

1 分钟69#海外 AI 资讯#AI in practice#Artificial Intelligence
AI&大模型
智东西10/17

梁圣又掀桌了!DeepSeek新模型斩杀GPT-6,许愿式编程时代降临精选导读

智东西 作者 | 杨京丽 编辑 | 李水青 智东西10月17日报道,刚刚,DeepSeek V4.1 Flash更新至V4.1-Flash-1017版本,官方网页端、App与API已同步更新至新模型,模型权重也已在HuggingFace和ModelScope开放下载。DeepSeek还向公众开放自研搜索引擎的Search(搜索)与Fetch(抓取)API,两个接口均定价为5元/千次。 此次更新距离该模型发布仅过去一个多月。9月10日,…

5 分钟66#智能硬件#芯片
AI&大模型
The Decoder10/17

AI models hide mistakes in their own work reports, but one sentence in your prompt can help精选导读

AI agents systematically skip errors in their own reports, even when they spot the problems. In a study by Google Research, MIT, and Harvard, GPT-5.5 left out negative results in 198 out of 200 paper abstracts. The arti…

1 分钟69#海外 AI 资讯#AI research#Artificial Intelligence
AI&大模型
量子位10/17

科技企业出海,为啥我推荐你去迪拜?精选导读

暂无摘要

1 分钟63#大模型#国内动态#资讯
AI&大模型
The Decoder10/17

Mathematicians say OpenAI doesn't understand its own Navier-Stokes proof精选导读

OpenAI submitted a 166-page solution to the Navier-Stokes problem but apparently doesn't fully understand its own proof. Several groups of experts dispute the claimed methodology, saying the actual approach differs from…

1 分钟74#海外 AI 资讯#AI research#Artificial Intelligence
AI&大模型
MarkTechPost10/17

DeepSeek Just Released V4.1-Flash-1017, Scores 77.0 on DeepSWE, Opens Search API精选导读

DeepSeek has just released DeepSeek-V4.1-Flash-1017, a new checkpoint of its V4.1-Flash model. It is live on the DeepSeek website, app and API. DeepSeek claims stronger scores, faster task completion and shorter outputs…

14 分钟64#研究进展#模型发布#Agentic AI
AI&大模型
Latent Space10/17

[AINews] High Taste Testers Ep. 1精选导读

a quiet day lets us pilot a new show

30 分钟52#AI 工程
AI&大模型
MarkTechPost10/17

20 Jev Alternatives: The Decision Models Taking On TypeSafe in 2026精选导读

TypeSafe launched Jev on September 15, 2026. It is a decision model, not a chatbot. You define the allowed answers, and Jev returns a label, a score or a yes/no probability. In about 4 weeks, at least 20 Jev alternative…

30 分钟66#研究进展#模型发布#Agentic AI