[Bugfix] Fix Typo from NVFP4 Refactor (#31977)
Signed-off-by: Robert Shaw <robshaw@redhat.com> Co-authored-by: Robert Shaw <robshaw@redhat.com>
This commit is contained in:
co-authored by
Robert Shaw
parent
6ebe34d6fa
commit
0fa8dd24d2
+8
@@ -0,0 +1,8 @@
|
|||||||
|
model_name: "RedHatAI/Qwen3-30B-A3B-NVFP4"
|
||||||
|
accuracy_threshold: 0.88
|
||||||
|
num_questions: 1319
|
||||||
|
num_fewshot: 5
|
||||||
|
server_args: "--enforce-eager --max-model-len 8192 --data-parallel-size 2 --enable-expert-parallel --all2all-backend deepep_low_latency"
|
||||||
|
env:
|
||||||
|
VLLM_USE_FLASHINFER_MOE_FP4: "1"
|
||||||
|
VLLM_FLASHINFER_MOE_BACKEND: "masked_gemm"
|
||||||
+8
@@ -0,0 +1,8 @@
|
|||||||
|
model_name: "nvidia/Qwen3-30B-A3B-NVFP4"
|
||||||
|
accuracy_threshold: 0.88
|
||||||
|
num_questions: 1319
|
||||||
|
num_fewshot: 5
|
||||||
|
server_args: "--enforce-eager --max-model-len 8192 --data-parallel-size 2 --enable-expert-parallel --all2all-backend deepep_low_latency"
|
||||||
|
env:
|
||||||
|
VLLM_USE_FLASHINFER_MOE_FP4: "1"
|
||||||
|
VLLM_FLASHINFER_MOE_BACKEND: "masked_gemm"
|
||||||
+8
@@ -0,0 +1,8 @@
|
|||||||
|
model_name: "nvidia/Qwen3-30B-A3B-NVFP4"
|
||||||
|
accuracy_threshold: 0.88
|
||||||
|
num_questions: 1319
|
||||||
|
num_fewshot: 5
|
||||||
|
server_args: "--enforce-eager --max-model-len 8192 --data-parallel-size 2 --enable-expert-parallel"
|
||||||
|
env:
|
||||||
|
VLLM_USE_FLASHINFER_MOE_FP4: "1"
|
||||||
|
VLLM_FLASHINFER_MOE_BACKEND: "throughput"
|
||||||
@@ -1,8 +1,10 @@
|
|||||||
|
Qwen3-30B-A3B-NvFp4-CT-fi-cutedsl-deepep-ll.yaml
|
||||||
|
Qwen3-30B-A3B-NvFp4-ModelOpt-fi-cutedsl-deepep-ll.yaml
|
||||||
|
Qwen3-30B-A3B-NvFp4-CT-fi-cutlass.yaml
|
||||||
|
Qwen3-30B-A3B-NvFp4-ModelOpt-fi-cutlass.yaml
|
||||||
Qwen3-30B-A3B-Fp8-AutoFp8-deepgemm-deepep-ht.yaml
|
Qwen3-30B-A3B-Fp8-AutoFp8-deepgemm-deepep-ht.yaml
|
||||||
Qwen3-30B-A3B-Fp8-AutoFp8-deepgemm-deepep-ll.yaml
|
Qwen3-30B-A3B-Fp8-AutoFp8-deepgemm-deepep-ll.yaml
|
||||||
Qwen3-30B-A3B-Fp8-AutoFp8-deepgemm.yaml
|
Qwen3-30B-A3B-Fp8-AutoFp8-deepgemm.yaml
|
||||||
Qwen3-30B-A3B-Fp8-CT-Block-deepgemm-deepep-ht.yaml
|
Qwen3-30B-A3B-Fp8-CT-Block-deepgemm-deepep-ht.yaml
|
||||||
Qwen3-30B-A3B-Fp8-CT-Block-deepgemm-deepep-ll.yaml
|
Qwen3-30B-A3B-Fp8-CT-Block-deepgemm-deepep-ll.yaml
|
||||||
Qwen3-30B-A3B-Fp8-CT-Block-deepgemm.yaml
|
Qwen3-30B-A3B-Fp8-CT-Block-deepgemm.yaml
|
||||||
Qwen3-30B-A3B-NvFp4-CT-fi-cutlass.yaml
|
|
||||||
Qwen3-30B-A3B-NvFp4-ModelOpt-fi-cutlass-fi-a2av.yaml
|
|
||||||
|
|||||||
@@ -0,0 +1 @@
|
|||||||
|
Qwen3-30B-A3B-NvFp4-CT-marlin.yaml
|
||||||
@@ -456,7 +456,7 @@ def prepare_nvfp4_moe_layer_for_fi_or_cutlass(
|
|||||||
NvFp4MoeBackend.VLLM_CUTLASS,
|
NvFp4MoeBackend.VLLM_CUTLASS,
|
||||||
NvFp4MoeBackend.FLASHINFER_CUTLASS,
|
NvFp4MoeBackend.FLASHINFER_CUTLASS,
|
||||||
NvFp4MoeBackend.FLASHINFER_TRTLLM,
|
NvFp4MoeBackend.FLASHINFER_TRTLLM,
|
||||||
NvFp4MoeBackend.FLASHINFER_TRTLLM,
|
NvFp4MoeBackend.FLASHINFER_CUTEDSL,
|
||||||
]
|
]
|
||||||
|
|
||||||
# Reorder [w1, w3] to [w3, w1] for FI NVFP4 MoE kernels.
|
# Reorder [w1, w3] to [w3, w1] for FI NVFP4 MoE kernels.
|
||||||
|
|||||||
Reference in New Issue
Block a user