deepseek-ai/DeepGEMM

Health v2 · maintenance only

Adoption not part of the score

7738 stars · 1216 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

DeepGEMM is a high-performance CUDA BLAS kernel library for NVIDIA tensor cores, providing FP8, FP4, and BF16 GEMMs plus fused MoE and other LLM computation primitives with runtime JIT compilation. It is designed to be clean and simple, matching or exceeding expert-tuned libraries on Hopper (SM90) and Blackwell (SM100) GPUs.

Use cases

• speed up fp8 gemm kernels for llm inference on h100
• efficient moe fused kernels with overlapped communication
• learn nvidia gpu kernel optimization techniques
• bf16 matrix multiplication on sm90 gpus
• integrate fast gemm into a training framework
• mqa scoring kernels for lightning indexer

When to choose

• you need near-peak tensor core GEMM performance on SM90/SM100 NVIDIA GPUs
• you want a small, readable CUDA codebase for studying kernel optimization
• you need FP8/FP4 GEMMs or fused MoE kernels for LLM workloads without heavy template frameworks

When to avoid

• you need BLAS on non-NVIDIA hardware or pre-Hopper GPUs
• you need a general-purpose BLAS with broad routine coverage beyond GEMM-style operations
• you require a stable API with long-term compatibility guarantees

Facets

library · maturity active

machine-learning gpu-computing llm-inference llm-training benchmarking deep-learning large-language-models gpu-computing machine-learning developer-tools python cpp cuda gemm blas tensor-cores fp8 moe jit-compilation kernel-optimization linux gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="deepseek-ai/DeepGEMM")