Installation

Prior to installing OLMo-core you should install PyTorch according to the official instructions specific to your operating system and hardware.

Then you can install OLMo-core from PyPI with:

pip install ai2-olmo-core

There are a number of optional dependencies that must be installed to use certain functionality as well, including:

• flash-attn, ring-flash-attn, and TransformerEngine for the corresponding attention backends.

• Liger-Kernel for a low-memory “fused-linear” loss implementation.

• torchao for float8 training (see olmo_core.float8).

• grouped_gemm for dropless mixture-of-experts (MoE) models (see olmo_core.nn.moe).