Stories by gdiamos

LLM-Deflate: Extracting LLMs into Datasets

AMD MI300X Memcpy Peer Deep Dive

Craylm: Open-source unified LLM training and inference for R1

Notes on Building LLMs in 2024

LLM Inference at the Memory Wall

LLMs hallucinate in critical enterprise scenarios

Training LLMs in the Wild Wild West – AMD multi-node

Open source ICD11 Mistral trained on AMD GPUs [video]

Good old fashion ML is dying

AI startup Lamini bets future on AMD's Instinct GPUs

How to use a LLM to classify text

Laminify: Instantly classify data with Lamini and Llama 2

Cramming an LLM platform into one container

1.109B times faster serving of finetuned LLMs

Specializing LLMs on Databricks

MLPerf Releases Inference Results