Update vLLM-fork version to v0.8.5.post1+Gaudi-1.21.3

.github/env/_build_image.sh

@@ -3,4 +3,4 @@
 # SPDX-License-Identifier: Apache-2.0
 
 export VLLM_VER=v0.10.0
-export VLLM_FORK_VER=v0.6.6.post1+Gaudi-1.20.0
+export VLLM_FORK_VER=v0.8.5.post1+Gaudi-1.21.3

AgentQnA/docker_image_build/build.yaml

@@ -20,7 +20,7 @@ services:
   vllm-gaudi:
     build:
       context: vllm-fork
-      dockerfile: Dockerfile.hpu
+      dockerfile: docker/Dockerfile.hpu
     extends: agent
     image: ${REGISTRY:-opea}/vllm-gaudi:${TAG:-latest}
   vllm-rocm:

AgentQnA/tests/step1_build_images.sh

@@ -41,7 +41,7 @@ function build_agent_docker_image_gaudi_vllm() {
     get_genai_comps
 
     git clone https://github.com/HabanaAI/vllm-fork.git && cd vllm-fork
-    VLLM_FORK_VER=v0.6.6.post1+Gaudi-1.20.0
+    VLLM_FORK_VER=v0.8.5.post1+Gaudi-1.21.3
     git checkout ${VLLM_FORK_VER} &> /dev/null && cd ../
 
     echo "Build agent image with --no-cache..."

AudioQnA/docker_image_build/build.yaml

@@ -82,7 +82,7 @@ services:
   vllm-gaudi:
     build:
       context: vllm-fork
-      dockerfile: Dockerfile.hpu
+      dockerfile: docker/Dockerfile.hpu
     extends: audioqna
     image: ${REGISTRY:-opea}/vllm-gaudi:${TAG:-latest}
   vllm-rocm:

AudioQnA/tests/test_compose_on_gaudi.sh

@@ -27,7 +27,7 @@ function build_docker_images() {
 
     git clone https://github.com/HabanaAI/vllm-fork.git
     cd vllm-fork/
-    VLLM_FORK_VER=v0.6.6.post1+Gaudi-1.20.0
+    VLLM_FORK_VER=v0.8.5.post1+Gaudi-1.21.3
     echo "Check out vLLM tag ${VLLM_FORK_VER}"
     git checkout ${VLLM_FORK_VER} &> /dev/null && cd ../
 

ChatQnA/docker_image_build/build.yaml

@@ -123,7 +123,7 @@ services:
   vllm-gaudi:
     build:
       context: vllm-fork
-      dockerfile: Dockerfile.hpu
+      dockerfile: docker/Dockerfile.hpu
     extends: chatqna
     image: ${REGISTRY:-opea}/vllm-gaudi:${TAG:-latest}
   nginx:

ChatQnA/tests/test_compose_faqgen_on_gaudi.sh

@@ -24,7 +24,7 @@ function build_docker_images() {
     docker build --no-cache -t ${REGISTRY}/comps-base:${TAG} --build-arg https_proxy=$https_proxy --build-arg http_proxy=$http_proxy -f Dockerfile .
     popd && sleep 1s
     git clone https://github.com/HabanaAI/vllm-fork.git && cd vllm-fork
-    VLLM_FORK_VER=v0.6.6.post1+Gaudi-1.20.0
+    VLLM_FORK_VER=v0.8.5.post1+Gaudi-1.21.3
     git checkout ${VLLM_FORK_VER} &> /dev/null && cd ../
 
     echo "Build all the images with --no-cache, check docker_image_build.log for details..."

ChatQnA/tests/test_compose_guardrails_on_gaudi.sh

@@ -24,7 +24,7 @@ function build_docker_images() {
     docker build --no-cache -t ${REGISTRY}/comps-base:${TAG} --build-arg https_proxy=$https_proxy --build-arg http_proxy=$http_proxy -f Dockerfile .
     popd && sleep 1s
     git clone https://github.com/HabanaAI/vllm-fork.git && cd vllm-fork
-    VLLM_FORK_VER=v0.6.6.post1+Gaudi-1.20.0
+    VLLM_FORK_VER=v0.8.5.post1+Gaudi-1.21.3
     git checkout ${VLLM_FORK_VER} &> /dev/null && cd ../
 
     echo "Build all the images with --no-cache, check docker_image_build.log for details..."

ChatQnA/tests/test_compose_on_gaudi.sh

@@ -24,7 +24,7 @@ function build_docker_images() {
     docker build --no-cache -t ${REGISTRY}/comps-base:${TAG} --build-arg https_proxy=$https_proxy --build-arg http_proxy=$http_proxy -f Dockerfile .
     popd && sleep 1s
     git clone https://github.com/HabanaAI/vllm-fork.git && cd vllm-fork
-    VLLM_FORK_VER=v0.6.6.post1+Gaudi-1.20.0
+    VLLM_FORK_VER=v0.8.5.post1+Gaudi-1.21.3
     git checkout ${VLLM_FORK_VER} &> /dev/null && cd ../
 
     echo "Build all the images with --no-cache, check docker_image_build.log for details..."

ChatQnA/tests/test_compose_without_rerank_on_gaudi.sh

@@ -24,7 +24,7 @@ function build_docker_images() {
     docker build --no-cache -t ${REGISTRY}/comps-base:${TAG} --build-arg https_proxy=$https_proxy --build-arg http_proxy=$http_proxy -f Dockerfile .
     popd && sleep 1s
     git clone https://github.com/HabanaAI/vllm-fork.git && cd vllm-fork
-    VLLM_FORK_VER=v0.6.6.post1+Gaudi-1.20.0
+    VLLM_FORK_VER=v0.8.5.post1+Gaudi-1.21.3
     git checkout ${VLLM_FORK_VER} &> /dev/null && cd ../
 
     echo "Build all the images with --no-cache, check docker_image_build.log for details..."

CodeGen/docker_image_build/build.yaml

@@ -52,7 +52,7 @@ services:
   vllm-gaudi:
     build:
       context: vllm-fork
-      dockerfile: Dockerfile.hpu
+      dockerfile: docker/Dockerfile.hpu
     extends: codegen
     image: ${REGISTRY:-opea}/vllm-gaudi:${TAG:-latest}
   dataprep:

CodeGen/tests/test_compose_on_gaudi.sh

@@ -27,7 +27,7 @@ function build_docker_images() {
 
     # Download Gaudi vllm of latest tag
     git clone https://github.com/HabanaAI/vllm-fork.git && cd vllm-fork
-    VLLM_FORK_VER=v0.6.6.post1+Gaudi-1.20.0
+    VLLM_FORK_VER=v0.8.5.post1+Gaudi-1.21.3
     echo "Check out vLLM tag ${VLLM_FORK_VER}"
     git checkout ${VLLM_FORK_VER} &> /dev/null && cd ../
 

CodeTrans/docker_image_build/build.yaml

@@ -34,7 +34,7 @@ services:
   vllm-gaudi:
     build:
       context: vllm-fork
-      dockerfile: Dockerfile.hpu
+      dockerfile: docker/Dockerfile.hpu
     extends: codetrans
     image: ${REGISTRY:-opea}/vllm-gaudi:${TAG:-latest}
   nginx:

CodeTrans/tests/test_compose_on_gaudi.sh

@@ -26,7 +26,7 @@ function build_docker_images() {
     popd && sleep 1s
 
     git clone https://github.com/HabanaAI/vllm-fork.git && cd vllm-fork
-    VLLM_FORK_VER=v0.6.6.post1+Gaudi-1.20.0
+    VLLM_FORK_VER=v0.8.5.post1+Gaudi-1.21.3
     git checkout ${VLLM_FORK_VER} &> /dev/null && cd ../
 
     echo "Build all the images with --no-cache, check docker_image_build.log for details..."

DeepResearchAgent/tests/test_compose_on_gaudi.sh

@@ -26,7 +26,7 @@ function build_docker_images() {
     docker build --no-cache -t ${REGISTRY}/comps-base:${TAG} --build-arg https_proxy=$https_proxy --build-arg http_proxy=$http_proxy -f Dockerfile .
     popd && sleep 1s
     git clone https://github.com/HabanaAI/vllm-fork.git && cd vllm-fork
-    VLLM_FORK_VER=v0.6.6.post1+Gaudi-1.20.0
+    VLLM_FORK_VER=v0.8.5.post1+Gaudi-1.21.3
     git checkout ${VLLM_FORK_VER} &> /dev/null && cd ../
 
     echo "Build all the images with --no-cache, check docker_image_build.log for details..."

DocSum/docker_image_build/build.yaml

@@ -63,6 +63,6 @@ services:
   vllm-gaudi:
     build:
       context: vllm-fork
-      dockerfile: Dockerfile.hpu
+      dockerfile: docker/Dockerfile.hpu
     extends: docsum
     image: ${REGISTRY:-opea}/vllm-gaudi:${TAG:-latest}

DocSum/tests/test_compose_on_gaudi.sh

@@ -47,7 +47,7 @@ function build_docker_images() {
     popd && sleep 1s
 
     git clone https://github.com/HabanaAI/vllm-fork.git && cd vllm-fork
-    VLLM_FORK_VER=v0.6.6.post1+Gaudi-1.20.0
+    VLLM_FORK_VER=v0.8.5.post1+Gaudi-1.21.3
     git checkout ${VLLM_FORK_VER} &> /dev/null && cd ../
 
     echo "Build all the images with --no-cache, check docker_image_build.log for details..."

FinanceAgent/docker_compose/intel/hpu/gaudi/README.md

@@ -102,7 +102,7 @@ cd vllm-fork
 VLLM_VER=$(git describe --tags "$(git rev-list --tags --max-count=1)")
 echo "Check out vLLM tag ${VLLM_VER}"
 git checkout ${VLLM_VER}
-docker build --no-cache -f Dockerfile.hpu -t opea/vllm-gaudi:latest --shm-size=128g . --build-arg https_proxy=$https_proxy --build-arg http_proxy=$http_proxy
+docker build --no-cache -f docker/Dockerfile.hpu -t opea/vllm-gaudi:latest --shm-size=128g . --build-arg https_proxy=$https_proxy --build-arg http_proxy=$http_proxy
 ```
 
 ## Validate Services

FinanceAgent/tests/test_compose_on_gaudi.sh

@@ -96,9 +96,9 @@ function build_vllm_docker_image() {
     fi
     cd ./vllm-fork
 
-    VLLM_FORK_VER=v0.6.6.post1+Gaudi-1.20.0
+    VLLM_FORK_VER=v0.8.5.post1+Gaudi-1.21.3
     git checkout ${VLLM_FORK_VER} &> /dev/null
-    docker build --no-cache -f Dockerfile.hpu -t $VLLM_IMAGE --shm-size=128g . --build-arg https_proxy=$HTTPS_PROXY --build-arg http_proxy=$HTTP_PROXY
+    docker build --no-cache -f docker/Dockerfile.hpu -t $VLLM_IMAGE --shm-size=128g . --build-arg https_proxy=$HTTPS_PROXY --build-arg http_proxy=$HTTP_PROXY
     if [ $? -ne 0 ]; then
         echo "$VLLM_IMAGE failed"
         exit 1

VisualQnA/docker_image_build/build.yaml

@@ -46,6 +46,6 @@ services:
   vllm-gaudi:
     build:
       context: vllm-fork
-      dockerfile: Dockerfile.hpu
+      dockerfile: docker/Dockerfile.hpu
     extends: visualqna
     image: ${REGISTRY:-opea}/vllm-gaudi:${TAG:-latest}

VisualQnA/tests/test_compose_on_gaudi.sh

@@ -27,7 +27,7 @@ function build_docker_images() {
     popd && sleep 1s
 
     git clone https://github.com/HabanaAI/vllm-fork.git && cd vllm-fork
-    VLLM_FORK_VER=v0.6.6.post1+Gaudi-1.20.0
+    VLLM_FORK_VER=v0.8.5.post1+Gaudi-1.21.3
     git checkout ${VLLM_FORK_VER} &> /dev/null && cd ../
 
     service_list="visualqna visualqna-ui lvm nginx vllm-gaudi"

one_click_deploy/common/update_images.sh

@@ -123,7 +123,7 @@ get_service_list() {
 # Defines the configurations for cloning prerequisite repositories.
 # A generic build function will use these settings.
 
-VLLM_FORK_VER="v0.6.6.post1+Gaudi-1.20.0"
+VLLM_FORK_VER=v0.8.5.post1+Gaudi-1.21.3
 
 # Config for examples using vLLM v0.8.3
 declare -A VLLM_8_3_CONFIG=(