NeMo Framework Megatron Backend

NVIDIA

Container

NVIDIA

NeMo Framework Megatron Backend

NVIDIA NeMo™ framework Megatron backend supports pre-training, post-training, and reinforcement learning of LLMs and multi-modal generative AI models with state-of-the-art data processing, model training techniques, and flexible deployment options.

NVIDIA AI Enterprise Supported

Layer	Label		Created
fcbb1f2d42fa3037d0ced7a652011f5e6d6ddaed72a2817eae8ea1dea8dac3d5	CONFIG	`Entrypoint /opt/nvidia/nvidia_entrypoint.sh; WorkingDir /workspace; ExposedPorts 6006/tcp, 8888/tcp`	02/18/2025 9:19 PM UTC
3d4f8fd7ae3271ede5731896ae0b216ca697f46221d96881c872adb836b805e5	ARG	`NVIDIA_BUILD_REF`	02/18/2025 9:19 PM UTC
ca887158ae213227f59ac1877f074ce32814f9865b175a0fe8d0c8b9e63218fa	LABEL	`com.nvidia.build.id=134983853`	02/18/2025 9:19 PM UTC
64250a97d846e3847eb6af3ce0c8c65fcbad7e50afe4e1ad26e4f373183cbe60	ENV	`NVIDIA_BUILD_ID=134983853`	02/18/2025 9:19 PM UTC
d8cf58f481ef998af1d0712eff659c777093c28ff230a97f23fbb955b6e41dd1	ARG	`NVIDIA_BUILD_ID`	02/18/2025 9:19 PM UTC
e04caf96e9946af564b7e48fedfbee6466409bf9697e0561fca2a28e7fafa66e	RUN	NVIDIA_BIGNLP_VERSION= TRT_VER=10.8.0.24 CUDA_VER=12.8 CUDNN_VER=9.5.1.17-1 NCCL_VER= CUBLAS_VER=12.6.1.4-1 NVRTC_VER=12.6.85-1 TRT_LLM_COMMIT=2830e4e8099c0d7e6402494865b9cb7433f51eb5 REINSTALL_APEX=False APEX_COMMIT=b216eeee7dd91478745496919e9c0167cc4c41e3 MCORE_COMMIT=1e24413e3a3ea7eafc5fd66654cc6c93247f6f45 SRC_DIR=/src/tensorrt_llm CPP_BUILD_DIR=/src/tensorrt_llm/cpp/build TARGET_ARCH=x86 INSTALL_VLLM=True ALIGNER_COMMIT=r0.7.0 LAUNCHER_COMMIT=4abd481402adc8a061942c486eda6d71f19de718 DATA_CURATOR_COMMIT=r0.7.0 INSTALL_DATA_CURATOR=True MODELOPT_VERSION=0.21.0 NEMO_COMMIT=r2.2.0 INSTALL_RESIL=True RESIL_COMMIT=b6eb61dbf9fe272b1a943b1b0d9efdde99df0737 NEMO_RUN_COMMIT=451f781c67698582b7ba91ac86aac0e5b5c6ae52 PRE_COMPILE_QUANT_KERNELS=False chmod -R a+w /workspace	02/18/2025 9:19 PM UTC
d30efb69ce8863f6db88f27b8225a5c7ef426dbbcc749cddaf289a59419cab0a	WORKDIR	`/workspace`	02/18/2025 9:19 PM UTC
330c8e6b8e7fc35d91dd0cbc4db2bd03d7bc4f377648d7507d5a580ba8814ac3	RUN	NVIDIA_BIGNLP_VERSION= TRT_VER=10.8.0.24 CUDA_VER=12.8 CUDNN_VER=9.5.1.17-1 NCCL_VER= CUBLAS_VER=12.6.1.4-1 NVRTC_VER=12.6.85-1 TRT_LLM_COMMIT=2830e4e8099c0d7e6402494865b9cb7433f51eb5 REINSTALL_APEX=False APEX_COMMIT=b216eeee7dd91478745496919e9c0167cc4c41e3 MCORE_COMMIT=1e24413e3a3ea7eafc5fd66654cc6c93247f6f45 SRC_DIR=/src/tensorrt_llm CPP_BUILD_DIR=/src/tensorrt_llm/cpp/build TARGET_ARCH=x86 INSTALL_VLLM=True ALIGNER_COMMIT=r0.7.0 LAUNCHER_COMMIT=4abd481402adc8a061942c486eda6d71f19de718 DATA_CURATOR_COMMIT=r0.7.0 INSTALL_DATA_CURATOR=True MODELOPT_VERSION=0.21.0 NEMO_COMMIT=r2.2.0 INSTALL_RESIL=True RESIL_COMMIT=b6eb61dbf9fe272b1a943b1b0d9efdde99df0737 NEMO_RUN_COMMIT=451f781c67698582b7ba91ac86aac0e5b5c6ae52 PRE_COMPILE_QUANT_KERNELS=False if [ $PRE_COMPILE_QUANT_KERNELS = "True" ]; then python -c "import modelopt.torch.quantization.extensions as ext; ext.precompile()"; fi	02/18/2025 9:19 PM UTC
5681a6547896feb89749974023f0fc2f10cab052a37f750e3be8a15b123a08c7	ARG	`PRE_COMPILE_QUANT_KERNELS=False`	02/18/2025 9:19 PM UTC
146353b39edc6cc3c51bffc53dd0a30f09e7611780db31f38ca3b16aa37cd65f	RUN	NVIDIA_BIGNLP_VERSION= TRT_VER=10.8.0.24 CUDA_VER=12.8 CUDNN_VER=9.5.1.17-1 NCCL_VER= CUBLAS_VER=12.6.1.4-1 NVRTC_VER=12.6.85-1 TRT_LLM_COMMIT=2830e4e8099c0d7e6402494865b9cb7433f51eb5 REINSTALL_APEX=False APEX_COMMIT=b216eeee7dd91478745496919e9c0167cc4c41e3 MCORE_COMMIT=1e24413e3a3ea7eafc5fd66654cc6c93247f6f45 SRC_DIR=/src/tensorrt_llm CPP_BUILD_DIR=/src/tensorrt_llm/cpp/build TARGET_ARCH=x86 INSTALL_VLLM=True ALIGNER_COMMIT=r0.7.0 LAUNCHER_COMMIT=4abd481402adc8a061942c486eda6d71f19de718 DATA_CURATOR_COMMIT=r0.7.0 INSTALL_DATA_CURATOR=True MODELOPT_VERSION=0.21.0 NEMO_COMMIT=r2.2.0 INSTALL_RESIL=True RESIL_COMMIT=b6eb61dbf9fe272b1a943b1b0d9efdde99df0737 NEMO_RUN_COMMIT=451f781c67698582b7ba91ac86aac0e5b5c6ae52 sed -i '/^LayerId/s/^/#/' /usr/local/lib/python3.12/dist-packages/cv2/typing/__init__.py	02/18/2025 9:19 PM UTC