nemo-automodel | NVIDIA NGC

NVIDIA

nemo-automodel

Container

NVIDIA

nemo-automodel

NVIDIA NeMo™ AutoModel accelerates LLM and VLM training and fine‑tuning with PyTorch DTensor‑native SPMD, day‑0 Hugging Face support, and optimized parallelism from single‑ to multi‑node scale.

Layer	Label		Created
de64286db5d70556913bfca855bcd990bf1a62f71837671da48c76e49f1f5dce	CONFIG	`Entrypoint /opt/nvidia/nvidia_entrypoint.sh; WorkingDir /opt/Automodel; ExposedPorts 6006/tcp, 8888/tcp`	04/26/2026 8:06 AM UTC
e5cc548a5d72177b55600195ac631f604f12001e4f6e9251a514ab1836e179c4	ARG	`TARGETARCH=amd64`	04/26/2026 8:06 AM UTC
659e473658fe0cba0d0171a051c069908bb0958f3d18bbefba2fbd74dd6663fa	ARG	`RC_DATE=26.04`	04/26/2026 8:06 AM UTC
178cb90d5c9fec8f098c292345e2bcc7d8d6869dfa68e000ca3a69295450c178	LABEL	`com.nvidia.build.ref=2c14eae97962250fe8f28c8f76cc3ae95ec110cf`	04/26/2026 8:06 AM UTC
736465bd073d0bc0b0a7181ae7147a05aff159e0e2d74329a88dbfc2ae5107fb	ARG	`NVIDIA_BUILD_REF=2c14eae97962250fe8f28c8f76cc3ae95ec110cf`	04/26/2026 8:06 AM UTC
7814251cb012c8e125ce597d68911ca5a5f3ad01d723fdb5a7c6e03e96f64c8b	LABEL	`com.nvidia.build.id=305635159`	04/26/2026 8:06 AM UTC
ec7823530cd54a250c2e444af9651cc00862a964ad9868554d956e298c7bc891	ENV	`NVIDIA_BUILD_ID=305635159`	04/26/2026 8:06 AM UTC
8412df16642ba1ac86b15bfac2ff30436c03d304b25fa2a5dd0fea033fbeac61	ARG	`NVIDIA_BUILD_ID=305635159`	04/26/2026 8:06 AM UTC
656e561faf2fd9069687979ca8299f87aa33db6730a7f4bc0f1c900b9ebeac65	RUN	`INSTALL_TE=True TE_COMMIT=release_v2.11 DEEPEP_COMMIT=7febc6e25660af0f54d95dd781ecdcd62265ecca INSTALL_BITSANDBYTES=True BITSANDBYTES_COMMIT=0.49.2 INSTALL_UCCL_EP=False BASE_IMAGE=pytorch AUTOMODEL_INSTALL=all UV_SYNC_ARGS=--locked TARGETARCH=amd64 chmod +x /opt/venv/env.sh`	04/26/2026 8:06 AM UTC
65b6b72fb7bde210a158b2f9c390316451496dba2203f7a1093f78a64a19d52c	COPY	`<<EOF /opt/venv/env.sh`	04/26/2026 8:06 AM UTC