[pull] main from NVIDIA:main

constraints.txt

@@ -4,3 +4,7 @@
 urllib3>=2.6.3
 # WAR against https://github.com/advisories/GHSA-8rrh-rw8j-w5fx
 wheel>=0.46.2
+# WAR against https://github.com/advisories/GHSA-7gcm-g887-7qv7
+protobuf>=6.33.5
+# WAR against https://github.com/advisories/GHSA-6mq8-rvhq-8wgg
+aiohttp>=3.13.3

examples/disaggregated/slurm/benchmark/run_benchmark_nv_sa.sh

@@ -191,6 +191,20 @@ for concurrency in ${concurrency_list}; do
         --percentile-metrics "ttft,tpot,itl,e2el" \
         $([ "${streaming}" = "false" ] && echo "--non-streaming")
 
+    # Print failed request count (consistent with non-nv_sa benchmark format)
+    python - "${output_dir}/result.json" <<-'PYEOF'
+	import json
+	import sys
+
+	try:
+	    with open(sys.argv[1], encoding="utf-8") as f:
+	        d = json.load(f)
+	    failed = d["num_prompts"] - d["completed"]
+	    print(f"Total failed requests: {failed}")
+	except (OSError, json.JSONDecodeError, KeyError) as exc:
+	    print(f"WARNING: failed to read request counts from {sys.argv[1]}: {exc}", file=sys.stderr)
+	PYEOF
+
     echo "Benchmark with concurrency ${concurrency} done"
     do_process_all_logs ${log_path}/ ${log_path}/concurrency_${concurrency} "log"
 done

jenkins/current_image_tags.properties

@@ -13,7 +13,7 @@
 #     images are adopted from PostMerge pipelines, the abbreviated commit hash is used instead.
 IMAGE_NAME=urm.nvidia.com/sw-tensorrt-docker/tensorrt-llm
 
-LLM_DOCKER_IMAGE=urm.nvidia.com/sw-tensorrt-docker/tensorrt-llm:pytorch-25.12-py3-x86_64-ubuntu24.04-trt10.14.1.48-skip-tritondevel-202602011118-10901
-LLM_SBSA_DOCKER_IMAGE=urm.nvidia.com/sw-tensorrt-docker/tensorrt-llm:pytorch-25.12-py3-aarch64-ubuntu24.04-trt10.14.1.48-skip-tritondevel-202602011118-10901
-LLM_ROCKYLINUX8_PY310_DOCKER_IMAGE=urm.nvidia.com/sw-tensorrt-docker/tensorrt-llm:cuda-13.1.0-devel-rocky8-x86_64-rocky8-py310-trt10.14.1.48-skip-tritondevel-202602011118-10901
-LLM_ROCKYLINUX8_PY312_DOCKER_IMAGE=urm.nvidia.com/sw-tensorrt-docker/tensorrt-llm:cuda-13.1.0-devel-rocky8-x86_64-rocky8-py312-trt10.14.1.48-skip-tritondevel-202602011118-10901
+LLM_DOCKER_IMAGE=urm.nvidia.com/sw-tensorrt-docker/tensorrt-llm:pytorch-25.12-py3-x86_64-ubuntu24.04-trt10.14.1.48-skip-tritondevel-202603051044-11898
+LLM_SBSA_DOCKER_IMAGE=urm.nvidia.com/sw-tensorrt-docker/tensorrt-llm:pytorch-25.12-py3-aarch64-ubuntu24.04-trt10.14.1.48-skip-tritondevel-202603051044-11898
+LLM_ROCKYLINUX8_PY310_DOCKER_IMAGE=urm.nvidia.com/sw-tensorrt-docker/tensorrt-llm:cuda-13.1.0-devel-rocky8-x86_64-rocky8-py310-trt10.14.1.48-skip-tritondevel-202603051044-11898
+LLM_ROCKYLINUX8_PY312_DOCKER_IMAGE=urm.nvidia.com/sw-tensorrt-docker/tensorrt-llm:cuda-13.1.0-devel-rocky8-x86_64-rocky8-py312-trt10.14.1.48-skip-tritondevel-202603051044-11898

requirements-dev.txt

@@ -36,7 +36,7 @@ opentelemetry-api>=1.26.0
 opentelemetry-exporter-otlp>=1.26.0
 opentelemetry-semantic-conventions-ai>=0.4.1
 fuzzywuzzy==0.18.0
-aiperf==0.3.0
+aiperf==0.4.0
 nanobind>=2.9.0
 nixl==0.8.0
 hf-transfer==0.1.9

requirements.txt

@@ -30,7 +30,7 @@ nvidia-modelopt[torch]~=0.37.0
 # torch 2.9.1+cu130 depends on nvidia-nccl-cu13==2.27.7
 nvidia-nccl-cu13>=2.27.7,<=2.28.9
 nvidia-cuda-nvrtc
-transformers==4.57.1
+transformers==4.57.3
 prometheus_client
 prometheus_fastapi_instrumentator
 pydantic>=2.9.1