English(EN) Alibaba Opens Qwen3.8-2.4T-A95B Weights as First Qwen-Max-Class MoE

阿里巴巴发布 Qwen3.8-2.4T-A95B MoE 模型开放权重

作者 PulseAugur 编辑部 · [1 个来源] · 2026-09-10 04:59

阿里巴巴已发布其 Qwen3.8-2.4T-A95B 模型的开放权重。该混合专家（MoE）模型拥有 2.4 万亿总参数，其中 950 亿为激活参数。它采用了混合注意力机制，专为长上下文、智能体（agentic）工作负载设计，并附带了 SageMaker HyperPod 和 vLLM 的配置指南。 AI

影响此次发布提供了一个强大的、具有开放权重的 MoE 模型，有望加速智能体 AI 和长上下文应用领域的研究与开发。

排序理由前沿实验室模型发布，开放权重。[lever_c_demoted from frontier_release: ic=1 ai=1.0]

在 Pandaily 阅读 →

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

本文如何被排名

Signal score: 67 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Significant; 前沿实验室模型发布，开放权重。[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]

Pandaily TIER_1 English(EN) · [email protected] (Pandaily) · 2026-09-10 04:59

阿里巴巴开放 Qwen3.8-2.4T-A95B 权重，首个 Qwen-Max 级别 MoE

Alibaba's Qwen team published open weights for Qwen3.8-2.4T-A95B, a 2.4-trillion-parameter MoE with 95 billion active parameters. Hybrid attention and long context target agentic workloads, with same-period SageMaker HyperPod and vLLM recipes.

本文如何被排名

报道来源 [1]

阿里巴巴开放 Qwen3.8-2.4T-A95B 权重，首个 Qwen-Max 级别 MoE

相关实体

相关话题