PulseAugur
中
实时 22:38:24
English(EN) Fourth Paradigm just bypassed Nvidia entirely to train two new foundation models on Huawei Ascend silicon. The standout is Phai-A1: an enterprise coding and age

Fourth Paradigm 在华为硬件上训练 AI 模型,绕过英伟达

Fourth Paradigm 开发了两个新的基础模型,包括 Phai-A1,利用华为昇腾芯片,绕过了英伟达硬件。Phai-A1 是一个面向企业的模型,专为编码和代理任务设计,拥有百万级 token 的上下文窗口。值得注意的是,它可以在单块消费级 RTX 3090 上进行推理,展示了本土化中国硬件的能力。 AI

影响 展示了非英伟达硬件在训练先进 AI 模型方面的可行性,可能使 AI 基础设施格局更加多元化。

排序理由 Frontier-lab 模型发布,附带系统卡。[lever_c_demoted from frontier_release: ic=1 ai=1.0]

在 Mastodon — mastodon.social 阅读 →

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

Fourth Paradigm 在华为硬件上训练 AI 模型,绕过英伟达

本文如何被排名

Signal score
17 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Significant
Frontier-lab 模型发布,附带系统卡。[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
Breaking (< 6h)
Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

  1. Mastodon — mastodon.social TIER_1 English(EN) · asiaai ·

    Fourth Paradigm 仅用华为昇腾芯片就训练出了两个新的基础模型,完全绕过了英伟达。其中最突出的是 Phai-A1:一个企业级编码和年龄

    Fourth Paradigm just bypassed Nvidia entirely to train two new foundation models on Huawei Ascend silicon. The standout is Phai-A1: an enterprise coding and agent model with a native 1-million-token context window that can run inference on a single consumer RTX 3090. A massive pr…