[Hardware][AMD][CI] Use Triton-based AITER MHA for LM Eval Qwen-3.5 Models Tests (#46691)

Signed-off-by: Matthew Wong <Matthew.Wong2@amd.com>
This commit is contained in:
Matt
2026-06-25 17:08:04 -05:00
committed by GitHub
parent a2e8ec3d52
commit 27da2a2ac4
3 changed files with 9 additions and 6 deletions
@@ -3,7 +3,6 @@ accuracy_threshold: 0.89
tolerance: 0.03
num_questions: 1319
num_fewshot: 5
startup_max_wait_seconds: 3600
server_args: >-
--max-model-len 4096
--tensor-parallel-size 2
@@ -11,3 +10,4 @@ server_args: >-
--moe-backend aiter
env:
VLLM_ROCM_USE_AITER: "1"
ENABLE_CK: "0" # Avoid AITER CK-based MHA JIT compilation to save time