开源大模型 Llama 4 的翻车还在持续发酵。 4 月 8 日，作为大语言模型「权威榜单之一」的 Chatbot Arena（民间俗称「大模型竞技场」）发布了一则语气 ...

使用微信扫码将网页分享到微信 昨天，Meta Llama 4 就这么突然地发布了。 纸面参数很高大上，号称原生多模态 MOE 模型，击败 DeepSeek V3，还有 2 万亿 ...

【新智元导读】Llama 4家族周末突袭，实属意外。这场AI领域的「闪电战」不仅带来了两款全新架构的开源模型，更揭示了一个惊人事实：苹果Mac设备 ...

Meta has debuted the first two models in its Llama 4 family, its first to use mixture of experts tech. A Saturday post from the social media giant announced the release of two models: ...

In the past few years, the AI world has shifted from a culture of open collaboration to one dominated by closely guarded ...

This is a developing situation, and a surprising one for Meta, whose AI reputation has improved with each new Llama release before this one.

The company was also concerned that Llama 4 was less capable than OpenAI's models in conducting humanlike voice conversations, the report added. Meta plans to spend as much as $65 billion this year to ...

Two of the models are now available but not its most powerful Llama 4 model, which Meta teased would be released at a later ...

AI benchmarking is critical to determine performance, but results can be irrelevant to enterprise workflows; enterprise ...