deepseek-ai/DeepGEMM
FP8 matrix multiplication library for NVIDIA Hopper, supports dense and MoE GEMMs
View on index · View in 3D Map
Related repos
- MLA decoding kernel for Hopper GPUs
- CUDA C++ templates for matrix multiplication and linear algebra
- DeepEP — communication library for Mixture-of-Experts and expert parallelism, with all-to-all GPU kernels and low-precision support
- Toolkit for optimizing and deploying deep learning models across CPUs, GPUs, and NPUs
- run local LLMs fast on consumer GPUs with flexible quantization
- toolkit for building and optimizing AI agent workflows
- n8n node for DeepSeek AI automations
- GPU-optimized library for training massive transformer models at scale
- GPU-accelerated evolutionary multiobjective optimization library
- open-source GPU cluster manager for AI models
