⬢github first seen 12 h ago, last 1 min ago, peak #9

DeepSeek releases DeepGEMM GPU kernel library

Original: deepseek-ai/DeepGEMM

DeepSeek's AI research arm has released DeepGEMM, an open-source library of clean and efficient BLAS kernels for GPUs, written in CUDA. The project is climbing developer rankings as engineers examine its performance optimizations, which are relevant to training and running large AI models efficiently on limited hardware.

Why now: Developers are closely watching DeepSeek's open-source tooling for efficient AI computation, and a fast GPU math library has broad practical appeal.

DeepSeekDeepGEMMGitHubNVIDIA CUDA

Open on github →

Rank over time, top of the chart is #1. 51 snapshots from 12 h ago to 1 min ago.

Evidence

• deepseek-ai/DeepGEMM · deepseek-ai · 16

  DeepGEMM: clean and efficient BLAS kernel library on GPU

API: https://socialmediatrends-api.osmike.com/v1/trends/1227357