mha

MySQL Proxy using Java NIO based on Sharding SQL,Calcite ,simple and fast

E-Mail Header Analyzer

Medical image processing in Python

基于Docker的MySQL MHA集群

Decoding Attention is specially optimized for MHA, MQA, GQA and MLA using CUDA core for the decoding stage of LLM inference.

Performance of the C++ interface of flash attention and flash attention v2 in large language model (LLM) inference scenarios.

Automatic failover mysql with consul

Preprocessing 3D medical images and image archives —geared towards prostate cancer detection in MRI.

Switch MySQL replication with custom scripts and MHA, also support proxysql

A community-driven, in-depth guide for the free-to-play game My Hero Ultra Rumble.

MHA, MQA, GQA, MLA 相关原理及简要实现

The MHA module handles installing, configuring, and running MHA for MySQL.

array-api based simple volume renderer

One of the best anime bots out there ;)

A community-driven, in-depth guide for the free-to-play game My Hero Ultra Rumble.

A code deep-dive on one of the key innovations from Deepseek - Multihead Latent Attention (MLA)

Teaching them to play together

This mod adds Tsuyu Asui from My Hero Acadamia into minetest

⚡ Optimize attention mechanisms with FlashMLA, a library of advanced sparse and dense kernels for DeepSeek models, improving performance and efficiency.

A GPT-style autoregressive language model built from scratch in PyTorch - no AutoModelForCausalLM, no GPT2LMHeadModel, no pretrained Transformer implementation. Implements token embeddings, positional embeddings, causal self-attention, grouped-query attention (GQA), Transformer decoder blocks, AdamW optimization, checkpointed GPU training.