NVIDIA
NVIDIA
Triton Inference Server
Container
NVIDIA
NVIDIA
Triton Inference Server

Triton Inference Server is an open source software that lets teams deploy trained AI models from any framework, from local or cloud storage and on any GPU- or CPU-based infrastructure in the cloud, data center, or embedded devices.

LayerLabelCreated
561aa04746e78d95565cc994f04797777bbfb1da603ff1aadd902e9ca9a70293CONFIG
Entrypoint /opt/nvidia/nvidia_entrypoint.sh; WorkingDir /opt/tritonserver
02/13/2026 9:38 PM UTC
14422ee652582207b4b190cd274fc8465cca16c5b791ec35cae49540fe6ca9caLABEL
com.amazonaws.sagemaker.capabilities.multi-models=true
02/13/2026 9:38 PM UTC
9d4d4fa1cc958fca27caba86d913e60aca31df36714c905fc7fe9a0b35221a0fLABEL
com.amazonaws.sagemaker.capabilities.accept-bind-to-port=true
02/13/2026 9:38 PM UTC
5fde205db33400e549a2ed216d9b7686cefe1129ab3f548bfbd58bc99a6c4fd7RUN
TRITON_VERSION=2.66.0 TRITON_CONTAINER_VERSION=26.02 pip3 install -r python/openai/requirements.txt
02/13/2026 9:38 PM UTC
45223d5fc1f0c31147af85661e858bb025fbbd9cf119aee9b5f86b92ac824fb3RUN
TRITON_VERSION=2.66.0 TRITON_CONTAINER_VERSION=26.02 find /opt/tritonserver/python -maxdepth 1 -type f -name "tritonserver-*.whl" | xargs -I {} pip install --upgrade {}[all] &&
  find /opt/tritonserver/python -maxdepth 1 -type f -name "tritonfrontend-*.whl" | xargs -I {} pip install --upgrade {}[all]
02/13/2026 9:38 PM UTC
1da9585a0b3c61e982ae3e53038a6d3c9a99c83c986db4a494a6f0c2410ffc51COPY
--chown=1000:1000 NVIDIA_Deep_Learning_Container_License.pdf .
02/13/2026 9:37 PM UTC
7bfd0bb8ad90c8be29c393df9b6c0c7043d25a52e5da51737d3429adac3325caWORKDIR
/opt/tritonserver
02/13/2026 9:37 PM UTC
fdf7eb059d34775733bc279729da241c7c582457898fb45a5903439727584d68COPY
--chown=1000:1000 build/install tritonserver
02/13/2026 9:37 PM UTC
578cdaa12125efc7cc040ebc25ecfd5779354183593dc5324bf58c88e957a0edWORKDIR
/opt
02/13/2026 9:37 PM UTC
994c34ec0599f41cbaf33e5c82a532c0e373e35d742cd90372f8498e95b8fa9aLABEL
com.nvidia.build.ref=90e8ae01b64a8511866ce5d876eab9d6179c8eee
02/13/2026 9:37 PM UTC
...