NVIDIA
NVIDIA
NeMo Framework Megatron Backend
Container
NVIDIA
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.

  • LayerLabelCreated
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4LABEL
    com.nvidia.build.ref=
    03/10/2024 7:52 PM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ARG
    NVIDIA_BUILD_REF
    03/10/2024 7:52 PM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4LABEL
    com.nvidia.build.id=80741402
    03/10/2024 7:52 PM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ENV
    NVIDIA_BUILD_ID=80741402
    03/10/2024 7:52 PM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ARG
    NVIDIA_BUILD_ID
    03/10/2024 7:52 PM UTC
    sha256:c065e287b6a1b63df8cc04d7a1cba8723af0d7ef015224136ecff7b3e2ce48e4RUN
    chmod -R a+w /workspace
    03/10/2024 7:52 PM UTC
    sha256:4f4fb700ef54461cfa02571ae0db9a0dc1e0cdb5577484a6d75e68dc38e8acc1WORKDIR
    /workspace
    03/10/2024 7:52 PM UTC
    sha256:9ed494f224ddaa88b13c36891603406a0f5931154d59c57c7ecc514cc8fa24adRUN
    TCNN_CUDA_ARCHITECTURES="52,60,61,70,75,80,86,89,90" pip install --no-cache-dir git+https://github.com/NVlabs/tiny-cuda-nn@6f018a9cd1b369bcb247e1d539968db8e48b2b3f#subdirectory=bindings/torch
    03/10/2024 7:52 PM UTC
    sha256:4f18d5d43c46042f8089a5d6a7deeec66d516b75a0ec155f67c801a59cb69ef5RUN
    cd /tmp &&
      git clone https://github.com/ashawkey/stable-dreamfusion.git &&
      cd stable-dreamfusion &&
      git checkout 5550b91862a3af7842bb04875b7f1211e5095a63 &&
      find . -type f -name 'setup.py' -exec sed -i 's/c++14/c++17/g' {} + &&
      pip install --no-cache-dir ./raymarching &&
      pip install --no-cache-dir ./shencoder &&
      pip install --no-cache-dir ./freqencoder &&
      TORCH_CUDA_ARCH_LIST="6.0 6.1 7.0 7.5 8.0 8.6 9.0+PTX" pip install --no-cache-dir ./gridencoder &&
      cd /opt &&
      rm -rf /tmp/stable-dreamfusion
    03/10/2024 7:05 PM UTC
    sha256:b1138e766af05f00dd5a38bf5658ad6ca00edbabb7372dedc9d28939d3387780RUN
    pip install --no-cache-dir tritonclient==2.38.0 kornia==0.6 Pillow==9.3.0 webdataset==0.2.48 git+https://github.com/openai/CLIP.git@main#egg=clip -e git+https://github.com/CompVis/taming-transformers.git@master#egg=taming-transformers --src /opt accelerate seaborn datasets prefetch_generator diffusers==0.19.3 torchdiffeq torchsde jiwer editdistance pyannote.core pyannote.metrics addict yapf basicsr open_clip_torch torch-ema trimesh PyMCubes xatlas imageio pymeshlab nerfacc==0.5.3 einops_exts git+https://github.com/NVlabs/nvdiffrast.git@c5caf7bdb8a2448acc491a9faa47753972edd380
    03/10/2024 6:39 PM UTC
    ...

    NVIDIA uses cookies to improve your experience on our web site. We and our third-party partners also use cookies and other tools to collect and record information you provide as well as information about your interactions with our websites for performance improvement, analytics, and to assist in marketing efforts. By clicking "Accept All", you consent to our use of cookies and other tools as described in our Cookie Policy. You can manage your cookie settings by clicking on "Manage Settings." By continuing to use this site or by clicking one of the buttons below, you agree to our Terms of Service (which contains important waivers). Please see our Privacy Policy for more information on our privacy practices.