Repository navigation
Conversation
sufubao
marked this pull request as draft
November 12, 2026 04:45
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
支持 Qwen3.8-Flash-Next/Qwen4Exp,接入模型、tokenizer、QSA、n-gram/PLE 和 MTP,并将 GDN/KDA ReplaySSM 与混合注意力状态管理整理到同一 PR。覆盖 history、accepted-state、页缓存、CPU/PD/DP 转移和 MTP 提交。
补齐 causal-convolution 的可选后端和本地 Triton fallback,支持缺少可加载
sgl_kernel的运行环境;保留 packed prefill、前缀状态、padding、非连续缓存和环形游标语义,并修复 SGL update 原地写入但不返回 Tensor 的适配。验证:H200 上 ReplaySSM fold/history 40 项、execution/acceptance 57 项回归通过;本次卷积数值、状态与 CUDA Graph 回归 114 项通过。相关 Python 通过 Black、Flake8。增加固定请求全文的 AgentX 控制回放:统一逐轮预热、缓存重置、SSE 用量审计和三轮波动门槛;相关回归 33 项通过,并验证 SGLang 缓存重置及阶段间空闲等待。H100 完整模型 ReplaySSM AgentX 尚无可用吞吐结论;公开数据回放不代表真实业务验收。