xgrammar

xgrammar

CONTAINERS IMAGES RUN BUILD

CONTAINERS

xgrammar:0.1.15
Requires	L4T ['>=34.1.0']
Dependencies	build-essential pip_cache:cu126 python cmake ninja cuda:12.6 cudnn numpy onnx pytorch:2.8 torchvision torchaudio huggingface_hub rust transformers
Dockerfile	Dockerfile
Notes	https://github.com/mlc-ai/xgrammar.git

xgrammar:0.1.18
Requires	L4T ['>=34.1.0']
Dependencies	build-essential pip_cache:cu126 python cmake ninja cuda:12.6 cudnn numpy onnx pytorch:2.8 torchvision torchaudio huggingface_hub rust transformers
Dockerfile	Dockerfile
Notes	https://github.com/mlc-ai/xgrammar.git

xgrammar:0.1.19
Requires	L4T ['>=34.1.0']
Dependencies	build-essential pip_cache:cu126 python cmake ninja cuda:12.6 cudnn numpy onnx pytorch:2.8 torchvision torchaudio huggingface_hub rust transformers
Dockerfile	Dockerfile
Notes	https://github.com/mlc-ai/xgrammar.git

xgrammar:0.1.20
Requires	L4T ['>=34.1.0']
Dependencies	build-essential pip_cache:cu126 python cmake ninja cuda:12.6 cudnn numpy onnx pytorch:2.8 torchvision torchaudio huggingface_hub rust transformers
Dockerfile	Dockerfile
Notes	https://github.com/mlc-ai/xgrammar.git

xgrammar:0.1.21
Aliases	xgrammar
Requires	L4T ['>=34.1.0']
Dependencies	build-essential pip_cache:cu126 python cmake ninja cuda:12.6 cudnn numpy onnx pytorch:2.8 torchvision torchaudio huggingface_hub rust transformers
Dependants	cosmos-reason1 dynamo:0.3.2 genai-bench:0.1.0 l4t-dynamo llama-factory sgl-kernel:0.2.3 sglang:0.4.4 sglang:0.4.6 sglang:0.4.9 vllm:0.7.4 vllm:0.8.4 vllm:0.9.0 vllm:0.9.2 vllm:0.9.3 vllm:v0.8.5.post1
Dockerfile	Dockerfile
Notes	https://github.com/mlc-ai/xgrammar.git

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 xgrammar)

# or if using 'docker run' (specify image and mounts/ect)
sudo docker run --runtime nvidia -it --rm --network=host xgrammar:36.4.0

^{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 xgrammar)

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

jetson-containers run $(autotag xgrammar) 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 xgrammar

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