PulseAugur
中
实时 11:21:59
English(EN) A Sovereign, Open-Source Foundation Model for German and English

开源 Soofi S 30B-A3B 模型发布，支持德语和英语

作者 PulseAugur 编辑部 · [3 个来源] ·

研究人员推出 Soofi S 30B-A3B，一个专为德语和英语设计的新型开源基础模型。该模型采用混合 Mamba-Transformer 架构，每个 token 只激活其 300 亿参数中的 30 亿，这为长上下文应用提供了显著的吞吐量优势。Soofi S 在约 27 万亿 token 上进行了训练，并展现出与更大密集模型相媲美的性能，尤其在代码生成任务上表现出色，超越了其他欧洲主权模型。该模型在慕尼黑的 Deutsche Telekom 德国工业人工智能云基础设施上开发，并将以宽松的开放访问条款发布，包括权重、代码和详细的数据账目。 AI

影响 这个开源模型可以加速德语和英语人工智能应用的开发和部署，特别是那些需要长上下文和高并发的应用。

排序理由 该集群描述了一篇详细介绍开源基础模型的新研究论文。

在 Hugging Face Daily Papers 阅读 →

AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →

开源 Soofi S 30B-A3B 模型发布，支持德语和英语

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
Research
该集群描述了一篇详细介绍开源基础模型的新研究论文。
Source corroboration
3 independent sources
Multiple independent publishers reporting the same story raises confidence that it's real and newsworthy.
Topics
model release, paper, 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
52 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.
Coverage growth since scoring
+1 source(s) since last score
New sources have picked up this story since our last re-score. Score will update on the next scoring pass.

完整方法见我们的编辑标准。

报道来源 [3]

  1. arXiv cs.AI TIER_1 English(EN) · The Soofi-Team, :, Benedikt Droste, David Fitzek, Ruben H\"arle, Lukas Helff, Maximilian Idahl, Alex Jude, Abbas Goher Khan, Maurice Kraus, Timm Ruland, Richard Rutmann, Sebastian Sztwiertnia, Markus Frey, Daniil Gurgurov, Jan Pfister, Tom R\"ohr, Sebas… ·

    面向德语和英语的、主权的、开源的基础模型

    arXiv:2607.09424v1 Announce Type: cross Abstract: We present Soofi S 30B-A3B, a sovereign, open-source Mixture-of-Experts (MoE) hybrid Mamba Transformer foundation model for German and English. Its hybrid design activates only 3B of 30B parameters per token and keeps the inferenc…

  2. arXiv cs.AI TIER_1 English(EN) · Max Lübbering ·

    面向德语和英语的、主权的、开源的基础模型

    We present Soofi S 30B-A3B, a sovereign, open-source Mixture-of-Experts (MoE) hybrid Mamba Transformer foundation model for German and English. Its hybrid design activates only 3B of 30B parameters per token and keeps the inference cache near-constant as context grows, giving it …

  3. Hugging Face Daily Papers TIER_1 English(EN) ·

    面向德语和英语的、主权的、开源的基础模型

    We present Soofi S 30B-A3B, a sovereign, open-source Mixture-of-Experts (MoE) hybrid Mamba Transformer foundation model for German and English. Its hybrid design activates only 3B of 30B parameters per token and keeps the inference cache near-constant as context grows, giving it …

相关实体

相关话题