Deepseek announced the open sourcing of DeepGEMM on the third day of Open Source Week. DeepGEMM is a library designed specifically for concise and efficient FP8 general matrix multiplication (GEMM), with fine-grained scaling capabilities, as proposed in DeepSeek-V3. It supports both regular and Mixture of Experts (MoE) grouped GEMMs. The library is written in CUDA, does not require compilation during installation, and compiles all kernels at runtime using a lightweight Just-In-Time (JIT) module. The FP8GEMM library provides support for V3/R1 training and inference.