PulseAugur
实时 17:53:31
English(EN) Qwen3.8-Flash-Next: A New Architecture, Towards Ultimate Cost-Efficiency https://qwen.ai/blog?id=qwen3.8-flash-next # HackerNews # Tech # AI

阿里巴巴推出 Qwen3.8-Flash 模型,获广泛合作伙伴支持

阿里巴巴的 Qwen 推出了其 Qwen3.8-Flash 模型,可在 Qwen Cloud 上使用,API 使用价格具有竞争力。该模型也可通过 OpenRouter 访问,支持编码助手和代理工作流等各种应用。此外,Qwen3.8-Flash-Next 已获得 SGLangvLLM 等关键合作伙伴的零日支持,并已在 NVIDIAAMD 硬件上确认兼容。这种新架构旨在实现极致的成本效益,也可通过 Ollama 获取。 AI

影响 通过有竞争力的定价加速多模态模型在编码和代理工作流中的应用。

排序理由 前沿实验室模型发布,附带系统卡。

在 Mastodon — sigmoid.social 阅读 →

AI 生成摘要 · Google Gemini · 来自 20 个来源。 我们如何撰写摘要 →

阿里巴巴推出 Qwen3.8-Flash 模型,获广泛合作伙伴支持

本文如何被排名

Signal score
1 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Frontier Release
前沿实验室模型发布,附带系统卡。
Source corroboration
20 independent sources
Strong cross-source corroboration — multiple independent publishers covered this within the clustering window.
Topics
model release, product, 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
2 days old
Coverage has settled into its steady-state source set.
Coverage growth since scoring
+3 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 [20]

  1. X — Qwen (Alibaba) TIER_1 English(EN) · Alibaba_Qwen ·

    Qwen3.8-Flash 在 @qwen_cloud 上:每100万输入令牌0.15美元,每100万输出令牌0.47美元,缓存命中仅需0.016美元/100万。☁️ 快来试试吧!👇

    Qwen3.8-Flash on @qwen_cloud: $0.15/1M input tokens, $0.47/1M output tokens, and just $0.016/1M on cache hits. ☁️ Come give it a try! 👇

  2. X — Qwen (Alibaba) TIER_1 English(EN) · Alibaba_Qwen ·

    Qwen3.8-Flash 现已在 @OpenRouter 上线!⚡ 编码助手、代理工作流、长视频理解,所有功能均可通过一次 API 调用实现。

    Qwen3.8-Flash is now live on @OpenRouter! ⚡ Coding assistants, agentic workflows, long-video understanding, all one API call away.

  3. X — Qwen (Alibaba) TIER_1 English(EN) · Alibaba_Qwen ·

    非常感谢 @sgl_project 的零日支持!🙌 Qwen3.8-Flash-Next 今天已准备好使用 SGLang 进行部署。

    Big thanks to @sgl_project for the day-0 support! 🙌 Qwen3.8-Flash-Next is ready to deploy with SGLang today.

  4. X — Qwen (Alibaba) TIER_1 English(EN) · Alibaba_Qwen ·

    太棒了!🥳 感谢 @vllm_project 在 NVIDIA 和 AMD 上从第一天起就运行了 Qwen3.8-Flash-Next。

    Amazing! 🥳 Thanks @vllm_project for getting Qwen3.8-Flash-Next running on NVIDIA and AMD from day 0.

  5. X — Qwen (Alibaba) TIER_1 English(EN) · Alibaba_Qwen ·

    非常感谢 @sgl_project 的零日支持!🙌 Qwen3.8-Flash-Next 今天已准备好使用 SGLang 进行部署。

    Big thanks to @sgl_project for the day-0 support! 🙌 Qwen3.8-Flash-Next is ready to deploy with SGLang today.

  6. Mastodon — sigmoid.social TIER_1 English(EN) · [email protected] ·

    Qwen3.8-Flash-Next 智能、性能与价格分析 https:// artificialanalysis.ai/models/q wen3-8-flash-next 评论: https:// news.ycombinator.co

    Qwen3.8-Flash-Next Intelligence, Performance and Price Analysis https:// artificialanalysis.ai/models/q wen3-8-flash-next Comments: https:// news.ycombinator.com/item?id=4 9461862 # HackerNews # Qwen3 .8 # Flash # Next # Intelligence # Performance # Price # Analysis # AI # Techno…

  7. Mastodon — sigmoid.social TIER_1 English(EN) · [email protected] ·

    Qwen3.8-Flash-Next 智能、性能与价格分析 https:// artificialanalysis.ai/models/q wen3-8-flash-next # ai

    Qwen3.8-Flash-Next Intelligence, Performance and Price Analysis https:// artificialanalysis.ai/models/q wen3-8-flash-next # ai

  8. Mastodon — sigmoid.social TIER_1 English(EN) · [email protected] ·

    在 NVIDIA GB300 NVL72 上使用 Qwen3.8-Flash-Next 进行 Agentic Coding 实验 #AgenticAI #AgenticArtificialIntelligence #A

    https://www. europesays.com/3217680/ Experiment with Qwen3.8-Flash-Next on NVIDIA GB300 NVL72 for Agentic Coding # AgenticAI # AgenticArtificialIntelligence # AI # ArtificialIntelligence

  9. Mastodon — sigmoid.social TIER_1 English(EN) · [email protected] ·

    🚀 Ollama 新模型发布!🚀 模型:qwen3.8-flash-next 🔗 https://ollama.com/library/qwen3.8-flash-next #Ollama #AI #LLM #Gemini #GeminiAPI

    🚀 New Ollama Model Release! 🚀 Model: qwen3.8-flash-next 🔗 https:// ollama.com/library/qwen3.8-fla sh-next # Ollama # AI # LLM # Gemini # GeminiAPI

  10. Mastodon — sigmoid.social TIER_1 English(EN) · [email protected] ·

    Qwen3.8-Flash-Next:新架构,迈向极致成本效益 https://qwen.ai/blog?id=qwen3.8-flash-next # ai

    Qwen3.8-Flash-Next: A New Architecture, Towards Ultimate Cost-Efficiency https:// qwen.ai/blog?id=qwen3.8-flash- next # ai

  11. Mastodon — sigmoid.social TIER_1 English(EN) · [email protected] ·

    Qwen3.8-Flash-Next:新架构,迈向极致性价比 https://qwen.ai/blog?id=qwen3.8-flash-next # HackerNews # Tech # AI

    Qwen3.8-Flash-Next: A New Architecture, Towards Ultimate Cost-Efficiency https://qwen.ai/blog?id=qwen3.8-flash-next # HackerNews # Tech # AI

  12. Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] ·

    Qwen3.8-Flash-Next https:// qwen.ai/blog?id=qwen3.8-flash- next # ai

    Qwen3.8-Flash-Next https:// qwen.ai/blog?id=qwen3.8-flash- next # ai

  13. Mastodon — fosstodon.org TIER_1 Deutsch(DE) · [email protected] ·

    RT @quxiaoyin: 我在 Claude Code 和 DeepSeek 的新代理平台上运行了相同的任务。一个花费了 150 美元,另一个仅花费了 2 美元

    RT @quxiaoyin: Ich habe dieselbe Aufgabe auf Claude Code und der neuen Agent-Plattform von DeepSeek ausgeführt. Die eine kostete 150 Dollar, die andere nur 2 Dollar. Heute starten wir https:// t.co/k6mrHjfQ3C (@agentskydev), das „OpenRouter für Agenten“ – eine API für Claude Code…

  14. Mastodon — fosstodon.org TIER_1 Deutsch(DE) · [email protected] ·

    RT @danielhanchen: Qwen 发布 Qwen3.8-Flash-Next,一款新的开源多模态 MoE 模型。💜 更多信息请访问 Arint.info #AI #MoE #Multimodal #OpenSource #

    RT @danielhanchen: Qwen kündigt Qwen3.8-Flash-Next an, ein neues Open-Weight-Multimodal-MoE-Modell. 💜 mehr auf Arint.info # AI # MoE # Multimodal # OpenSource # Qwen # UnslothAI # arint_info https://x.com/danielhanchen/status/2092222459550585019

  15. Mastodon — fosstodon.org TIER_1 Deutsch(DE) · [email protected] ·

    RT @UnslothAI: 现在您可以使用我们的笔记本免费微调 Qwen3.8-27B!🔥 更多信息请访问 Arint.info #AI #深度学习 #微调 #机器学习

    RT @UnslothAI: Jetzt kannst du Qwen3.8-27B kostenlos mit unserem Notebook feinabstimmen! 🔥 mehr auf Arint.info # AI # DeepLearning # FineTuning # MachineLearning # OpenSource # Qwen3 # arint_info https://x.com/UnslothAI/status/2092255772214509698

  16. Mastodon — fosstodon.org TIER_1 Deutsch(DE) · [email protected] ·

    @OliDietzel: 距离 # Qwen3 .8-Flash-Next 发布还有 8 小时 x.com/ModelScope2022… 更多信息请访问 Arint.info # AI # FlashNext # ModelScope # Qwen3 # Release

    @OliDietzel: Noch 8 Stunden bis zum Release von # Qwen3 .8-Flash-Next x.com/ModelScope2022… mehr auf Arint.info # AI # FlashNext # ModelScope # Qwen3 # Release # arint_info https://x.com/OliDietzel/status/2092503972552577279

  17. Mastodon — mastodon.social TIER_1 Deutsch(DE) · [email protected] ·

    RT @OrcaRouter: Qwen3.8 Flash Next. Uncensored. Now in NVFP4. 🐳 more on Arint.info # AI # HuggingFace # MachineLearning # NVFP4 # OpenSource # Qwen3 # arint_

    RT @OrcaRouter: Qwen3.8 Flash Next. Unzensiert. Jetzt in NVFP4. 🐳 mehr auf Arint.info # AI # HuggingFace # MachineLearning # NVFP4 # OpenSource # Qwen3 # arint_info https://x.com/OrcaRouter/status/2093233719473746065

  18. Mastodon — mastodon.social TIER_1 Deutsch(DE) · [email protected] ·

    RT @0xBakeer: Qwen3.8-Flash-Next 在单台 DGX Spark 上:每秒高达 97 个 token。更多信息请访问 Arint.info # AI # DGXSpark # MachineLearning # NLP # Open

    RT @0xBakeer: Qwen3.8-Flash-Next auf einer einzelnen DGX Spark: bis zu 97 Tokens pro Sekunde. mehr auf Arint.info # AI # DGXSpark # MachineLearning # NLP # OpenSource # Qwen3 # arint_info https://x.com/0xBakeer/status/2092709567503229015

  19. Mastodon — mastodon.social TIER_1 Deutsch(DE) · [email protected] ·

    RT @vllm_project: Qwen3.8-Flash-Next 来自 @AlibabaQwen 在 vLLM 中获得 Day-0 支持,已在 NVIDIA 和 AMD GPU 上验证。🎉 更多信息请访问 Arint.info # AI # AMD # Machin

    RT @vllm_project: Qwen3.8-Flash-Next von @AlibabaQwen hat Day-0-Support in vLLM, verifiziert auf NVIDIA- und AMD-GPUs. 🎉 mehr auf Arint.info # AI # AMD # MachineLearning # NVIDIA # Qwen # vLLM # arint_info https://x.com/vllm_project/status/2092600887873286157

  20. Mastodon — mastodon.social TIER_1 English(EN) · [email protected] ·

    Qwen3.8-Flash-Next https://simonwillison.net/2026/Aug/26/qwen38-flash-next/ # AI # LLM # Tech

    Qwen3.8-Flash-Next https://simonwillison.net/2026/Aug/26/qwen38-flash-next/ # AI # LLM # Tech