Hacker GitHub

← All repos

deepseek-ai/DeepGEMM

View on GitHub ↗ ★ 9,177 Cuda

DeepGEMM: clean and efficient BLAS kernel library on GPU

6 mentions on Hacker News by 5 people: 1 story, 5 comments · first 2025-02-26 · latest 2026-04-24

Mentions per year 6 mentions · busiest year 2025 (5)
20082026
AllStoriesComments NewestTopOldest
  1. …foundational library for LLMs https://github.com/deepseek-ai/TileKernels Others worth mentioning: https://github.com/deepseek-ai/DeepGEMM a competitive foundational library https://github.com/deepseek-ai/Engram…
    comment by alecco · 2026-04-24 · in DeepSeek v4
  2. And the yield bit modification: https://github.com/deepseek-ai/DeepGEMM/blob/main/deep_gemm/...
    comment by fulafel · 2025-02-26 · in DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling
  3. The secret is here https://github.com/deepseek-ai/DeepGEMM/blob/main/deep_gemm%...
    comment by azurezyq · 2025-02-26 · in DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling
  4. It seems mostly Python, which is nice: https://github.com/deepseek-ai/DeepGEMM/blob/main/deep_gemm/...
    comment by fulafel · 2025-02-26 · in DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling
  5. Github: https://github.com/deepseek-ai/DeepGEMM - Up to 1350+ FP8 TFLOPS on Hopper GPUs - No heavy dependency, as clean as a tutorial - Fully Just-In-Time compiled - Core logic at ~300 lines - yet outperforms expert-tuned kernels across…
    comment by helloericsf · 2025-02-26 · in DeepSeek Open Source DeepGEMM – FP8 GEMM Library(300 lines for 1350+ FP8 TFLOPS)