From 79626586e1c21f9167ce2b5f371f17bc63b427c7 Mon Sep 17 00:00:00 2001 From: tzhouam Date: Mon, 9 Mar 2026 08:02:25 +0000 Subject: [PATCH] Update GPU memory utilization in qwen3_omni.yaml from 0.6 to 0.45 to prevent out-of-memory errors during model execution. Signed-off-by: tzhouam --- tests/perf/stage_configs/qwen3_omni.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/perf/stage_configs/qwen3_omni.yaml b/tests/perf/stage_configs/qwen3_omni.yaml index 802f8dd24..fd5586b94 100644 --- a/tests/perf/stage_configs/qwen3_omni.yaml +++ b/tests/perf/stage_configs/qwen3_omni.yaml @@ -47,7 +47,7 @@ stage_args: model_arch: Qwen3OmniMoeForConditionalGeneration worker_type: ar scheduler_cls: vllm_omni.core.sched.omni_ar_scheduler.OmniARScheduler - gpu_memory_utilization: 0.6 + gpu_memory_utilization: 0.45 enforce_eager: false trust_remote_code: true engine_output_type: latent # Output codec codes for code2wav