[PD] Add PD support for hybrid model (Qwen3-Next, DeepSeek V3.2 Exp) (#10912)

Signed-off-by: Shangming Cai <csmthu@gmail.com> Co-authored-by: hzh0425 <hzh0425@apache.org> Co-authored-by: ZeldaHuang <hzm414167@alibaba-inc.com>
2025-10-16 09:59:14 +08:00
parent 97d857c096
commit 868403f642
13 changed files with 730 additions and 189 deletions
--- a/test/srt/run_suite.py
+++ b/test/srt/run_suite.py
@@ -163,6 +163,7 @@ suites = {
        TestFile("test_deepseek_v3_basic.py", 275),
        TestFile("test_deepseek_v3_mtp.py", 275),
        TestFile("test_disaggregation_different_tp.py", 600),
+        TestFile("test_disaggregation_hybrid_attention.py", 200),
        TestFile("test_disaggregation_pp.py", 140),
    ],
    "per-commit-4-gpu-b200": [