Stories by ashvardanian

DeepSpeed-Chat: Easy, Fast and Affordable RLHF Training of ChatGPT-Like Models

How long will it take to train GPT-4 on Fugaku supercomputer

Kids Use Hnswlib for HNSW

Oracle Database 23c Free Developer-Release for Linux X64 (OL8 or RHEL8)

Finetuning LLMs on a Single GPU Using Gradient Accumulation

Quantum Computers Break the Internet Starting Now

Apple Laying Off Employees Will Deal Significant Damage to Company Morale

The Nanotube Compiler and Framework

Intel Continues with More Big-Time Optimizations to the Linux Kernel

Show HN: Faster FastAPI with simdjson and io_uring on Linux 5.19

EXT4 Scores a Nice Direct I/O Performance Improvement with Linux 6.3

You don't need a thousand GPUs to train a Vision-Language Transformer for Search

OpenAssistant – ChatGPT's Open Alternative (We need your help)

Transcending Scaling Laws with 0.1% Extra Compute

The insert benchmark on ARM and x86 cloud servers

Dennis Whyte: Nuclear Fusion and the Future of Energy – Lex Fridman Podcast #353

-memory-safe C++ – Jim Radigan – CppCon 2022

ChatGPT Creator OpenAI Discussing Offer Valuing Company at $29B

Data2vec 2.0: Highly efficient self-supervised learning for vision, speech, text

ARM’s Scalable Vector Extensions: A Critical Look at SVE2 for Integer Workloads

5
6
7
8
9
10
11
12