PulseAugur
实时 22:46:56
English(EN) LongCat-2.0: How Meituan Trained a 1.6T-Parameter Coding Model Without a Single Nvidia GPU

美团在国产ASIC上训练1.6T编码模型,绕过英伟达

美团发布了LongCat-2.0,一个为代理编码任务设计的1.6万亿参数的混合专家模型。值得注意的是,该模型完全在国产中国AI ASIC上训练,避免了英伟达硬件。LongCat-2.0采用了一种新颖的稀疏注意力机制,可处理百万级token上下文,并在软件工程基准测试中展现出与领先的闭源模型相媲美的性能。 AI

影响 证明了在非英伟达硬件上训练大型模型的可能性,可能实现AI基础设施的多元化并降低成本。

排序理由 一家主要科技公司(美团)发布的新前沿模型,包含重要的基础设施和技术细节。[lever_c_demoted from frontier_release: ic=1 ai=1.0]

在 dev.to — LLM tag 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。 我们如何撰写摘要 →

美团在国产ASIC上训练1.6T编码模型,绕过英伟达

本文如何被排名

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Significant
一家主要科技公司(美团)发布的新前沿模型,包含重要的基础设施和技术细节。[lever_c_demoted from frontier_release: ic=1 ai=1.0]
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
model release, infra
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
47 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准

报道来源 [1]

  1. dev.to — LLM tag TIER_1 English(EN) · Prabhakar Chaudhary ·

    LongCat-2.0:美团如何在没有一块英伟达 GPU 的情况下训练出 1.6T 参数的编码模型

    <h1> LongCat-2.0: How Meituan Trained a 1.6T-Parameter Coding Model Without a Single Nvidia GPU </h1> <p>Meituan — best known in China as a food delivery and super-app company — quietly ran one of the most technically ambitious open-weight model releases of 2026. LongCat-2.0 is a…