From 82a38de2c4b6ee549ed636c50845aceff1368775 Mon Sep 17 00:00:00 2001
From: Aaron Rueth <rueth@google.com>
Date: Thu, 30 Oct 2025 22:44:35 +0000
Subject: [PATCH] Updated vllm version

---
 .../terraform/_shared_config/inference-ref-arch_variables.tf  | 4 ++--
 .../model-download/huggingface/download_wait.sh               | 2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/platforms/gke/base/use-cases/inference-ref-arch/terraform/_shared_config/inference-ref-arch_variables.tf b/platforms/gke/base/use-cases/inference-ref-arch/terraform/_shared_config/inference-ref-arch_variables.tf
index 80515abdb..a86da1c73 100644
--- a/platforms/gke/base/use-cases/inference-ref-arch/terraform/_shared_config/inference-ref-arch_variables.tf
+++ b/platforms/gke/base/use-cases/inference-ref-arch/terraform/_shared_config/inference-ref-arch_variables.tf
@@ -48,7 +48,7 @@ variable "ira_online_gpu_kubernetes_service_account_name" {
 }
 
 variable "ira_online_gpu_vllm_image_url" {
-  default     = "docker.io/vllm/vllm-openai:v0.10.1.1"
+  default     = "docker.io/vllm/vllm-openai:v0.11.0"
   description = "The URL for the GPU vLLM container image."
   type        = string
 }
@@ -72,7 +72,7 @@ variable "ira_online_tpu_max_diffusion_sdxl_image_url" {
 }
 
 variable "ira_online_tpu_vllm_image_url" {
-  default     = "docker.io/vllm/vllm-tpu:4c409cabc2c1c432ba670029990bd59e6bbf1479"
+  default     = "docker.io/vllm/vllm-tpu:v0.11.1"
   description = "The URL for the TPU vLLM container image."
   type        = string
 }
diff --git a/test/scripts/platforms/gke/base/use-cases/inference-ref-arch/model-download/huggingface/download_wait.sh b/test/scripts/platforms/gke/base/use-cases/inference-ref-arch/model-download/huggingface/download_wait.sh
index 23caa0556..62a34c109 100755
--- a/test/scripts/platforms/gke/base/use-cases/inference-ref-arch/model-download/huggingface/download_wait.sh
+++ b/test/scripts/platforms/gke/base/use-cases/inference-ref-arch/model-download/huggingface/download_wait.sh
@@ -31,7 +31,7 @@ kubectl_wait(){
   source "${ACP_REPO_DIR}/platforms/gke/base/use-cases/inference-ref-arch/terraform/_shared_config/scripts/set_environment_variables.sh"
 
   echo "Waiting for '${HF_MODEL_ID}'(${HF_MODEL_ID_HASH}) download" | ts "$(date +'%Y-%m-%d %H:%M:%S.%N %Z') [${HF_MODEL_ID}]"
-  kubectl --namespace=${huggingface_hub_downloader_kubernetes_namespace_name} wait job/${HF_MODEL_ID_HASH}-hf-model-to-gcs --for=condition=complete --timeout=14400s  | ts "$(date +'%Y-%m-%d %H:%M:%S.%N %Z') [${HF_MODEL_ID}]" &
+  kubectl --namespace=${huggingface_hub_downloader_kubernetes_namespace_name} wait job/${HF_MODEL_ID_HASH}-hf-model-to-gcs --for=condition=complete --timeout=14400s | ts "$(date +'%Y-%m-%d %H:%M:%S.%N %Z') [${HF_MODEL_ID}]" &
   kubectl --namespace=${huggingface_hub_downloader_kubernetes_namespace_name} wait job/${HF_MODEL_ID_HASH}-hf-model-to-gcs --for=condition=failed --timeout=14400s | ts "$(date +'%Y-%m-%d %H:%M:%S.%N %Z') [${HF_MODEL_ID}]" && exit 1 &
   wait -n && \
   pkill -f "kubectl --namespace=${huggingface_hub_downloader_kubernetes_namespace_name} wait job/${HF_MODEL_ID_HASH}-hf-model-to-gcs" || true