Brand new Co-Main Event podcast: Conor McGregor finds a way to see his injury as a good thing actually and DDP figures out one way to dissuade people from shooting on him (knee em in the damn dome). Plus more!

LLM Compressor v0.14.0 is out, and GPTQ just got its biggest speedup since launch. A new Triton kernel makes quantization ~15x faster end to end. Batching layers that share a shape pushes that to ~30x on some MoE workloads. Even the old eager path is 1.5-2x faster. Also new: expanded MSE/iMatrix observers that beat GPTQ for NVFP4 on internal benchmarks, REAP pruning with distributed DDP, and support for GLM 5.3 and Qwen3.8. Full release notes:

Build and train an LLM "from scratch" yourself and you truly understand what's happening inside 🛠️ A complete educational implementation that runs on a single GPU. Title: FareedKhan-dev/train-llm-from-scratch URL: 🛠️ Overview An educational repository that implements a Transformer from scratch in PyTorch, based on "Attention is All You Need." It promises you can train your own million- to billion-parameter LLM on a single GPU. ❓ Challenges Solved LLMs are ubiquitous, but hands-on chances to train one from scratch and understand its internals are rare. ・Just using off-the-shelf frameworks leaves the Transformer's mechanics opaque ・Learners needed an end-to-end resource spanning pretraining through post-training alignment 💡 Content & Structure It covers the entire LLM lifecycle. ・Data acquisition and preprocessing (from The Pile) ・Core Transformer architecture (embeddings, attention, feed-forward networks) ・Model training (with DDP for distributed processing) ・Post-training alignment: SFT, reward modeling, PPO, DPO, GRPO ・Text generation and inference Code is organized into src/models, scripts, data_loader, configs, and a Streamlit ui. The stack is PyTorch, tiktoken, HDF5, and NumPy. 🌍 Use Cases / Audience For developers and researchers who want hands-on understanding of LLM training — from those with limited GPUs (starting at 13M parameters) to those targeting multi-billion-parameter models on enterprise hardware. #LLM# #MachineLearning#

ENHYPEN WORLD TOUR ‘WALK THE LINE’ : FINAL THE CITY SEOUL 📅 2025.10.11 (SAT) – 2025.11.15 (SAT) 📍 Seoul, South Korea ENHYPEN SEOUL THE CITY takes over Seoul! Check out the full program at the link below. ▶️ #엔하이픈# #ENHYPEN# #EN_WORLDTOUR_WALKTHELINE# #EN_WALKTHELINE_FINAL# #EN_THECITY_SEOUL# 🌆 Media facade @ DDP @ Sejong Center for the Performing Arts @ Gwanghwamun Square Haechi Madang @ Banpo Bridge 💃 Random Play Dance @ Sinchon Car-Free Street @ Mangwon Seoul Battleship Park 🛍️ Pop-up Store @ 2F Lotte World Mall, Jamsil 📸 Photo Booth @ Photoism 🥨 F&B @ GABAEDO Myeongdong branch+6 @ FELT COFFEE Cheonggyecheon branch @ Knows Jongno Punggyeong branch+2 @ CHUNG KI WA TOWN Namyeong branch+3 @ Euljiro Jeokdang main branch Get ready for special moments with ENHYPEN! ✨