English(EN) Deploying Qwen3.8-2.4T-A95B on Amazon SageMaker HyperPod with vLLM

阿里巴巴发布 Qwen3.8-2.4T-A95B 作为开放权重前沿模型

作者 PulseAugur 编辑部 · [1 个来源] · 2026-09-09 22:26

阿里巴巴的 Qwen 团队发布了 Qwen3.8-2.4T-A95B，这是 Qwen-Max 级别模型的首次开放权重发布。该大型模型采用混合注意力架构，拥有 2.4 万亿参数，每个 token 激活 950 亿参数，专为要求严苛的智能体和推理任务而设计。此次发布包括开放权重，并展示了使用 vLLM 和 NVIDIA B300 Blackwell Ultra GPU 在 Amazon SageMaker HyperPod 上的部署，提供了一个与 OpenAI 兼容的端点。 AI

影响为开放权重前沿模型树立了新的标杆，有望加速复杂智能体任务的自托管 AI 部署。

排序理由阿里巴巴发布了带有开放权重和系统卡的 Frontier-lab 模型。[lever_c_demoted from frontier_release: ic=1 ai=1.0]

在 AWS Machine Learning Blog 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: 70 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Significant; 阿里巴巴发布了带有开放权重和系统卡的 Frontier-lab 模型。[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]

AWS Machine Learning Blog TIER_1 English(EN) · Dmitry Soldatkin · 2026-09-09 22:26

在 Amazon SageMaker HyperPod 上使用 vLLM 部署 Qwen3.8-2.4T-A95B

Learn how to deploy Qwen3.8-2.4T-A95B, a 2.4-trillion-parameter open-weight model, on Amazon SageMaker HyperPod with vLLM. This walkthrough covers cluster provisioning, NVFP4 quantization, and an OpenAI-compatible endpoint with built-in reasoning, tool calling, and native MTP spe…

本文如何被排名

报道来源 [1]

在 Amazon SageMaker HyperPod 上使用 vLLM 部署 Qwen3.8-2.4T-A95B

相关实体

相关话题