[Bug]: Wide speculative decoding verify (num_speculative_tokens ≥ 7) samples EOS at `prompt_len+8` on hybrid GDN models (Qwen3.5): silent ea
![[Bug]: Wide speculative decoding verify (num_speculative_tokens ≥ 7) samples EOS at `prompt_len+8` on hybrid GDN models (Qwen3.5): silent ea](https://www.chat-gpts.plus/wp-content/uploads/2026/10/61738-8d640711-768x403.jpg)
在 vLLM 启用投机解码(speculative decoding)跑 Qwen3.5 这类 hybrid GDN 模型时,可能在第 10 个 token 附近静默提前停止( finish_reason=stop ),输出在 EOS 行之前逐字节干净;Issue 讨论最终判定为 bf16 下 EO








