lmarena.ai(@lmarena_ai)

Dive into the Agent Arena leaderboard for more information: https://t.co/YQZuZQuoFU

8.5内容质量
Dive into the Agent Arena leaderboard for more information: https://t.co/YQZuZQuoFU

TL;DR · AI 摘要

Qwen3.8-Flash-Next在Agent Arena排行榜中表现优异,排名第七,具有显著的性能提升。

核心要点

  • Qwen3.8-Flash-Next在Agent Arena开放模型中排名第7,净提升+2.4%
  • 模型参数规模达125B,推理价格0.16美元/百万输入token
  • 性能指标显示Confirmed Success提升12.3%,无工具幻觉问题

结构提纲

按章节快速跳转。

  1. Qwen3.8-Flash-NextAgent Arena开放模型中排名第7,净提升2.4%。

  2. 模型采用MoE架构,参数规模125B,支持多模态处理。

  3. 通过QwenCloud API提供服务,输入token价格0.16美元/百万。

  4. Confirmed Success指标提升12.3%,无工具幻觉问题。

思维导图

用一张图看清主题之间的关系。

查看大纲文本(无障碍 / 无 JS 友好)
  • Qwen3.8-Flash-Next性能分析
    • 排行榜表现
      • 开放模型排名第7
      • 总体排名第24
    • 技术参数
      • 125B参数规模
      • MoE架构
    • 商业化信息
      • QwenCloud API
      • 输入价格0.16美元/百万

金句 / Highlights

值得收藏与分享的关键句。

#Qwen3.8-Flash-Next#Agent Arena#Alibaba Cloud#模型性能
打开原文

Arena.ai on X: "Dive into the Agent Arena leaderboard for more information: https://t.co/YQZuZQuoFU" / X

[](https://x.com/)

Arena.ai on X: "Dive into the Agent Arena leaderboard for more information: https://t.co/YQZuZQuoFU"

  • ![Image 1: Avatar](https://x.com/arena) Arena.ai @arena 23h Qwen3.8-Flash-Next by @Alibaba_Qwen has landed in Agent Arena, ranking #7 among open models (#24 overall) with +2.4% net improvement across 8.7K+ real-world agentic sessions! Among open models, it sits just behind DeepSeek V4 Flash (High) at #6 (+3% net improvement) and two Show more ![Image 2](https://x.com/arena/status/2094566204488962483/photo/1) [](https://x.com/arena/status/2094566204488962483/photo/1) ![Image 3](https://x.com/Alibaba_Qwen/status/2092591393424515114/photo/1) [](https://x.com/Alibaba_Qwen/status/2092591393424515114/photo/1) ![Image 4: @Alibaba_Qwen](https://x.com/Alibaba_Qwen)Qwen @Alibaba_Qwen Aug 26 ⚡Meet Qwen3.8-Flash, a multimodal MoE and an early preview of the Qwen4 architecture, now open-weight! The production version Qwen3.8-Flash will be available soon via QwenCloud API at just $ 0.16/1M input tokens and $ 0.47/1M output tokens. 125B parameters + 51B N-gram Show more 19 26 297 46K
  • ![Image 5: Avatar](https://x.com/arena) Arena.ai @arena 23h Qwen3.8-Flash-Next is #24 overall with +2.4% net improvement (#7 among open), and by signal: - Confirmed Success: +12.3% - Bash Recovery: +2.9% - Praise vs. Complaint: -1.6% - Steerability: -1.1% - No issues with Tool Hallucination ![Image 6](https://x.com/arena/status/2094566207794061800/photo/1) [](https://x.com/arena/status/2094566207794061800/photo/1) 2 4 13 5.7K
  • ![Image 7: Avatar](https://x.com/arena) Arena.ai @arena Dive into the Agent Arena leaderboard for more information: ![Image 8 Agent Arena | AI Agent Performance Leaderboard](https://t.co/YQZuZQuoFU)From arena.ai 11:21 PM · Aug 31, 20264.3K Views [](https://x.com/i/status/2094566210021278022) 1 4 1

Log in or sign up for X

See what’s happening and join the conversation

Continue with phoneContinue with Apple Continue with Google

or

Log in with username or email