tensorrt_llm

tensorrt_llm

CONTAINERS IMAGES RUN BUILD

CONTAINERS

tensorrt_llm:0.10.dev0
Aliases	tensorrt_llm
Requires	L4T ['==r36.*', '>=cu124']
Dependencies	build-essential cuda:11.4 cudnn python tensorrt numpy cmake onnx pytorch:2.2 torchvision huggingface_hub rust transformers cuda-python
Dockerfile	Dockerfile
Notes	The tensorrt-llm:builder container includes the C++ binaries under /opt

tensorrt_llm:0.10.dev0-builder
Aliases	tensorrt_llm:builder
Requires	L4T ['==r36.*', '>=cu124']
Dependencies	build-essential cuda:11.4 cudnn python tensorrt numpy cmake onnx pytorch:2.2 torchvision huggingface_hub rust transformers cuda-python
Dockerfile	Dockerfile
Notes	The tensorrt-llm:builder container includes the C++ binaries under /opt

tensorrt_llm:0.5
Aliases	tensorrt_llm
Requires	L4T ['==r36.*', '==cu122']
Dependencies	build-essential cuda:11.4 cudnn python tensorrt numpy cmake onnx pytorch:2.2 torchvision huggingface_hub rust transformers cuda-python
Dockerfile	Dockerfile
Notes	The tensorrt-llm:builder container includes the C++ binaries under /opt

tensorrt_llm:0.5-builder
Aliases	tensorrt_llm:builder
Requires	L4T ['==r36.*', '==cu122']
Dependencies	build-essential cuda:11.4 cudnn python tensorrt numpy cmake onnx pytorch:2.2 torchvision huggingface_hub rust transformers cuda-python
Dockerfile	Dockerfile
Notes	The tensorrt-llm:builder container includes the C++ binaries under /opt

RUN CONTAINER

To start the container, you can use jetson-containers run and autotag, or manually put together a docker run command:

# automatically pull or build a compatible container image
jetson-containers run $(autotag tensorrt_llm)

# or if using 'docker run' (specify image and mounts/ect)
sudo docker run --runtime nvidia -it --rm --network=host tensorrt_llm:35.2.1

^{jetson-containers run forwards arguments to docker run with some defaults added (like --runtime nvidia, mounts a /data cache, and detects devices)
autotag finds a container image that's compatible with your version of JetPack/L4T - either locally, pulled from a registry, or by building it.}

To mount your own directories into the container, use the -v or --volume flags:

jetson-containers run -v /path/on/host:/path/in/container $(autotag tensorrt_llm)

To launch the container running a command, as opposed to an interactive shell:

jetson-containers run $(autotag tensorrt_llm) my_app --abc xyz

You can pass any options to it that you would to docker run, and it'll print out the full command that it constructs before executing it.

BUILD CONTAINER

If you use autotag as shown above, it'll ask to build the container for you if needed. To manually build it, first do the system setup, then run:

jetson-containers build tensorrt_llm

The dependencies from above will be built into the container, and it'll be tested during. Run it with --help for build options.