PulseAugur
中
实时 07:09:30
English(EN) Congrats to @lightseekorg on the impressive TokenSpeed results! 🚀 Great to see Qwen3.8-Flash-Next reaching 685 output tokens/s per user on Nvida GPUs. Excited t

Qwen3.8-Flash-Next 在 NVIDIA GPU 上达到每秒 685 个 token

Lightseekorg 在 Qwen3.8-Flash-Next 模型上取得了新的基准测试结果,在使用 NVIDIA GB300 GPU 和 NVFP4 时,每用户每秒可输出 685 个 token。这一结果凸显了该模型的推理潜力和其在内核和部署基础设施方面取得的优化。阿里巴巴的 Qwen 团队祝贺 Lightseekorg 取得令人印象深刻的 TokenSpeed 结果,并鼓励进一步探索该模型的潜力。 AI

影响 展示了 Qwen 模型在推理速度方面的显著提升,可能影响部署成本和实时应用性能。

排序理由 现有模型的基准测试结果,并非新发布。 [lever_c_demoted from research: ic=1 ai=1.0]

在 X — Qwen (Alibaba) 阅读 →

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

Qwen3.8-Flash-Next 在 NVIDIA GPU 上达到每秒 685 个 token

本文如何被排名

Signal score
6 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
现有模型的基准测试结果,并非新发布。 [lever_c_demoted from research: 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. X — Qwen (Alibaba) TIER_1 English(EN) · Alibaba_Qwen ·

    恭喜 @lightseekorg 取得令人印象深刻的 TokenSpeed 结果!🚀 很高兴看到 Qwen3.8-Flash-Next 在 Nvidia GPU 上达到每位用户 685 输出 tokens/s。激动人

    Congrats to @lightseekorg on the impressive TokenSpeed results! 🚀 Great to see Qwen3.8-Flash-Next reaching 685 output tokens/s per user on Nvida GPUs. Excited to see the community unlock more of Qwen’s inference potential!