[Bugfix] Disable monolithic TRTLLM MoE for Renormalize routing (#37591) (#37605)

Signed-off-by: Vadim Gimpelson <vadim.gimpelson@gmail.com>

.buildkite/test_areas/lm_eval.yaml

@@ -45,6 +45,22 @@ steps:
   commands:
   - pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=configs/models-blackwell.txt
 
+- label: LM Eval Qwen3.5 Models (B200)
+  timeout_in_minutes: 120
+  device: b200
+  optional: true
+  num_devices: 2
+  source_file_dependencies:
+  - vllm/model_executor/models/qwen3_5.py
+  - vllm/model_executor/models/qwen3_5_mtp.py
+  - vllm/transformers_utils/configs/qwen3_5.py
+  - vllm/transformers_utils/configs/qwen3_5_moe.py
+  - vllm/model_executor/models/qwen3_next.py
+  - vllm/model_executor/models/qwen3_next_mtp.py
+  - vllm/model_executor/layers/fla/ops/
+  commands:
+  - pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=configs/models-qwen35-blackwell.txt
+
 - label: LM Eval Large Models (H200)
   timeout_in_minutes: 60
   device: h200

tests/evals/gsm8k/configs/Qwen3.5-35B-A3B-DEP2.yaml

@@ -0,0 +1,8 @@
+model_name: "Qwen/Qwen3.5-35B-A3B"
+accuracy_threshold: 0.86
+num_questions: 1319
+num_fewshot: 5
+server_args: >-
+  --max-model-len 4096
+  --data-parallel-size 2
+  --enable-expert-parallel

tests/evals/gsm8k/configs/Qwen3.5-35B-A3B-FP8-DEP2.yaml

@@ -0,0 +1,9 @@
+model_name: "Qwen/Qwen3.5-35B-A3B-FP8"
+accuracy_threshold: 0.86
+num_questions: 1319
+num_fewshot: 5
+server_args: >-
+  --max-model-len 4096
+  --data-parallel-size 2
+  --enable-expert-parallel
+  --kv-cache-dtype fp8

tests/evals/gsm8k/configs/models-qwen35-blackwell.txt

@@ -0,0 +1 @@
+Qwen3.5-35B-A3B-DEP2.yaml

vllm/model_executor/layers/fused_moe/experts/trtllm_fp8_moe.py

@@ -269,9 +269,16 @@ class TrtLlmFp8ExpertsMonolithic(TrtLlmFp8ExpertsBase, mk.FusedMoEExpertsMonolit
         weight_key: QuantKey | None,
         activation_key: QuantKey | None,
     ) -> bool:
-        """Monolithic kernels need to express router support."""
+        """Monolithic kernels need to express router support.
+        Renormalize/RenormalizeNaive are excluded: the monolithic kernel's
+        internal routing for these methods produces output uncorrelated
+        with the modular kernel's output and with Triton kernel's output
+        for Qwen3.5-35B-A3B-FP8.
+        See: https://github.com/vllm-project/vllm/issues/37591
+        """
         # NOTE(dbari): TopK routing could also be enabled, but need to validate models
         # NOTE(dbari): Default is not implemented and should not be enabled until it is
+
         if (weight_key, activation_key) in [
             (kFp8Static128BlockSym, kFp8Dynamic128Sym),
             (kMxfp8Static, kMxfp8Dynamic),
@@ -279,16 +286,12 @@ class TrtLlmFp8ExpertsMonolithic(TrtLlmFp8ExpertsBase, mk.FusedMoEExpertsMonolit
             # NOTE(rob): potentially allow others here. This is a conservative list.
             return routing_method in [
                 RoutingMethodType.DeepSeekV3,
-                RoutingMethodType.Renormalize,
-                RoutingMethodType.RenormalizeNaive,
             ]
         elif (weight_key, activation_key) == (kFp8StaticTensorSym, kFp8StaticTensorSym):
             # NOTE(dbari): as above, potentially allow others here.
             return routing_method in [
                 RoutingMethodType.DeepSeekV3,
                 RoutingMethodType.Llama4,
-                RoutingMethodType.Renormalize,
-                RoutingMethodType.RenormalizeNaive,
             ]
         else:
             raise ValueError("Unsupported quantization scheme.")