fix: mismatch GAP and PA version (#1386)

container/Dockerfile.sglang

@@ -10,7 +10,6 @@ ARG BASE_IMAGE_TAG="25.01-cuda12.8-devel-ubuntu24.04"
 ARG RELEASE_BUILD
 ARG RUNTIME_IMAGE="nvcr.io/nvidia/cuda"
 ARG RUNTIME_IMAGE_TAG="12.8.1-runtime-ubuntu24.04"
-ARG GENAI_PERF_VERSION=0.0.13
 
 # Define general architecture ARGs for supporting both x86 and aarch64 builds.
 #   ARCH: Used for package suffixes (e.g., amd64, arm64)
@@ -361,11 +360,6 @@ ENV LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/opt/nvidia/nvda_nixl/lib/x86_64-linux-gnu/
 ########################################
 FROM ci_minimum AS dev
 
-ARG GENAI_PERF_VERSION
-
-# Install genai-perf for benchmarking
-RUN uv pip install genai-perf==$GENAI_PERF_VERSION
-
 ENTRYPOINT ["/opt/nvidia/nvidia_entrypoint.sh"]
 
 CMD []

container/Dockerfile.tensorrt_llm

@@ -16,7 +16,6 @@
 ARG BASE_IMAGE="nvcr.io/nvidia/pytorch"
 ARG BASE_IMAGE_TAG="25.04-py3"
 ARG RELEASE_BUILD
-ARG GENAI_PERF_VERSION=0.0.13
 
 # Define general architecture ARGs for supporting both x86 and aarch64 builds.
 #   ARCH: Used for package suffixes (e.g., amd64, arm64)
@@ -163,11 +162,6 @@ RUN [ -f /etc/pip/constraint.txt ] && : > /etc/pip/constraint.txt || true && \
          "${TENSORRTLLM_PIP_WHEEL}" ; \
     fi
 
-ARG GENAI_PERF_VERSION
-
-# Install genai-perf for benchmarking
-RUN pip install genai-perf==$GENAI_PERF_VERSION
-
 # Install test dependencies
 RUN --mount=type=bind,source=./container/deps/requirements.test.txt,target=/tmp/requirements.txt \
     pip install --requirement /tmp/requirements.txt

container/Dockerfile.vllm

@@ -10,7 +10,6 @@ ARG BASE_IMAGE_TAG="25.01-cuda12.8-devel-ubuntu24.04"
 ARG RELEASE_BUILD
 ARG RUNTIME_IMAGE="nvcr.io/nvidia/cuda"
 ARG RUNTIME_IMAGE_TAG="12.8.1-runtime-ubuntu24.04"
-ARG GENAI_PERF_VERSION=0.0.13
 
 # Define general architecture ARGs for supporting both x86 and aarch64 builds.
 #   ARCH: Used for package suffixes (e.g., amd64, arm64)
@@ -452,11 +451,6 @@ ENV LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/opt/nvidia/nvda_nixl/lib/x86_64-linux-gnu/
 ########################################
 FROM ci_minimum AS dev
 
-ARG GENAI_PERF_VERSION
-
-# Install genai-perf for benchmarking
-RUN uv pip install genai-perf==$GENAI_PERF_VERSION
-
 ENTRYPOINT ["/opt/nvidia/nvidia_entrypoint.sh"]
 
 CMD []

container/deps/requirements.txt

@@ -16,6 +16,7 @@
 accelerate==1.6.0
 fastapi==0.115.6
 ftfy
+genai-perf==0.0.13
 grpcio-tools==1.66.0
 httpx
 kubernetes==32.0.1
@@ -36,6 +37,5 @@ sentencepiece
 tensorboard==2.19.0
 tensorboardX==2.6.2.2
 transformers
-tritonclient==2.53.0
 types-PyYAML
 uvicorn