跳到正文
𝕏 ModelScope(x.com)·· 5 天前精選品質評分74
重點導讀

長上下文模型常需要更多運算成本,Tencent Hy 推出開放權重模型 Hy4-flash,想讓長時間工作更省資源。

  • 它有 220B 引數,每個 token 只啟動 12B 引數,能理解圖片,並支援 1M-token 上下文。Tencent Hy 表示,在自家 WorkIndex 測試中,Hy4-flash 略勝較大的 Hy4 Preview,每項任務成本是 Hy4 Preview 的 1/13、Hy4 的 1/20。
  • 解碼速度超過每秒 150 tokens。模型也提供不同用途的版本,並支援 vLLM、SGLang、Transformers 和 OpenAI 相容 API。

AI 伴隨問答

自備 Key 深度分析

針對這篇報導有疑問?可直接點選快捷問題或向 AI 追問背景脈絡與技術細節:

推薦理由

Hy4-flash把長上下文、圖片理解與較低推論成本放進同一模型,文中也提供內部測試和部署資料,可用來瞭解它的定位。

ModelScope
ModelScope@ModelScope2022
·5 天前
𝕏

Tencent Hy has released Hy4-flash as an open-weight 220B MoE model that activates just 12B parameters per token, with native image understanding and 1M context.
🤖 https://t.co/5WWRWVRxRk

🏆 Built for lower-cost agent work: On Tencent’s internal WorkIndex, Hy4 flash slightly outperforms the much larger Hy4 Preview while cutting per-task cost to 1/13 of Hy4 Preview and 1/20 of Hy4.
⚡ Made to serve at scale: It exceeds 150 tokens per second in decoding, while its KV cache is just 1/24 the size of Hy3’s and 1/6 of Hy4’s.
🧠 Long context without the usual overhead: Hybrid attention, LatentMoE, Engram, and the built-in DFly2 drafter support 1M-token workflows with lower, more predictable decoding cost.
🖼️ Three checkpoints, different jobs: Choose Hy4 flash or Hy4 flash-FP8 for instruction following and reasoning, or Hy4 flash-base for continued pretraining and customization.
🛠️ Ready for common runtimes: Production recipes are available for vLLM and SGLang, with Transformers support and an OpenAI-compatible API workflow.
📜 Tencent Hy License applies.

推文來源 · X (Twitter)在 X 參與討論

讀者討論

理性交流 · 即時討論