模型发布/更新
-
小米发布并开源 MiMo-V2.6 系列,扩展强化学习规模迈向自我提升
Xiaomi MiMo 于今日正式发布并开源 MiMo-V2.6 系列模型,包含 Pro 与 Flash 两个原生全模态模型,称这是探索 RSI(递归自我改进)路径的关键一步。
-
xAI releases Grok 4.7, focused on coding and knowledge work
xAI released Grok 4.7, positioned as its strongest model for coding and knowledge work, using a larger base model than…
-
Xiaomi releases MiMo-V2.6 Pro and Flash omni-modal open-source models
Xiaomi MiMo released two omni-modal models, MiMo-V2.6 Pro and Flash, trained with scaled-up reinforcement learning.…
-
小米发布开放权重模型 MiMo-V2.6-Pro,登顶 Artificial Analysis 开放权重模型智能指数
小米(Xiaomi)发布开放权重模型 MiMo-V2.6-Pro,在 Artificial Analysis Intelligence Index 得分 46,超越前代 MiMo-V2.5-Pro 的 26,为开放权重模型中最高。
-
通义千问发布 Qwen-Image-2.1:7B 单检查点同时支持图像生成与编辑
通义千问发布 Qwen-Image-2.1,一个 7B 参数的原生图像生成与编辑模型,生成和编辑共用单一检查点,最多支持 10 张参考图。模型自带提示词增强 LLM,已集成 diffusers 和 ComfyUI,并在 Hugging F…
-
小米 MiMo-V2.6-Pro 登上 Code Arena: WebDev 榜约第10名,开源权重中约第3
小米发布 MiMo-V2.6-Pro 与 Flash 两款全模态模型,MiMo-V2.6-Pro 在 Code Arena: WebDev 榜以 1628 分(AutoEval)位列约第10名,开源权重(MIT 许可)中约第3,较 MiM…
产品发布/更新
-
Kimi releases Kimi Code Desktop 1.0 for macOS and Windows
Kimi (Moonshot AI) released Kimi Code Desktop 1.0, the official desktop client for Kimi Code, available simultaneously…
行业动态
-
British Columbia sues OpenAI, saying flagged ChatGPT activity was not referred to police before shooting
British Columbia sued OpenAI in California, saying flagged ChatGPT activity should have been reported to police before…
-
Amazon blocks Meta Muse agent from shopping for users, escalating dispute
Amazon blocked Meta's personal AI agent Muse from shopping on its website on users' behalf, saying it never authorized…
技巧与观点
-
马斯克称 Grok 4.7 使 xAI 在智能体编码领域位列第三
Elon Musk 引用 Artificial Analysis 评测称,Grok 4.7 使 xAI 在智能体编码上排名第三,仅次于 Anthropic 和 OpenAI。
-
Nathan Lambert analyzes the open-source model balance of power between China and the US
Nathan Lambert published an overview of the state of open-source models that he presented to the US Congress, noting…
-
Artificial Analysis evaluates Grok 4.7: intelligence index 46 enters top four, coding agent rises to No. 4
Artificial Analysis released its evaluation of Grok 4.7. Grok 4.7 (xhigh) scored 46 on the intelligence index, 2…
-
Digital Life Kazik interviews fansub and scanlation groups: the real situation of subtitle and manga groups in the AI era
The author interviewed members of several fansub and manga scanlation groups and found they do not resist AI. Eliza's…
-
Tomer Tunguz on AI-optimized if-then decisions: specialized decision-makers cut classification costs nearly 100-fold
In an article, Tomer Tunguz argues that the latest wave of AI is taking over the if-then decision primitive in…
-
Linear shares CI speedup: PR wait time falls from over 6 minutes to just over 5, per-test machine time roughly halved
Linear's engineering team detailed its CI optimization work: with its test suite nearly quadrupling since the start of…