# Qwen3.8-Flash-Next 发布，融合 Qwen4 架构创新

- 来源：SemiAnalysis (@SemiAnalysis_)
- 发布时间：2026-08-27 03:00
- AIHOT 分数：51
- AIHOT 链接：https://aihot.news/items/cmtah8b0e01kvrovuou1fz9lg
- 原文链接：https://x.com/SemiAnalysis_/status/2092688580111974648

## AI 摘要

阿里通义千问发布 Qwen3.8-Flash-Next，采用与即将推出的 Qwen4 相同的架构创新。包括 510 亿参数的 N-gram Embedding（可将嵌入表卸载至低速廉价 DRAM）、门控残差连接（GR），以及微块粒度选择上下文的 Qwen 稀疏注意力（QSA）。这些创新展示了中国开源模型在残差连接等方向的前沿探索。

## 正文

Congrats to @Alibaba_Qwen on the release of Qwen3.8-Flash-Next, using the same architecture innovations as their upcoming Qwen4 model! Such innovations include:

🟠 51-billion-param N-gram Embedding to look up a table with very little extra computation, which means the embedding table can be offloaded to slower & less expensive tiers of DRAM
🟠 Gated Residual (GR): it seems like a lot of Chinese labs are now innovating on the res connections, like Kimi's AttentionRes and DeepSeek's mHC
🟠 Qwen Sparse Attention (QSA): lightning indexer to select context at micro-block granularity

Glad to see great Chinese open innovations along with end-to-end model weights to show these innovations can compose well together!