Granite 4.1 LLMs: How They’re Built

A Dream of Spring for Open-Weight LLMs: 10 Architectures from Jan-Feb 2026

A Round Up And Comparison of 10 Open-Weight LLM Releases in Spring 2026

Consilium: When Multiple LLMs Collaborate

Recent Developments in LLM Architectures: KV Sharing, mHC, and Compressed Attention

From Gemma 4 to DeepSeek V4, How New Open-Weight LLMs Are Reducing Long-Context Costs

Making thousands of open LLMs bloom in the Vertex AI Model Garden

Controlling Reasoning Effort in LLMs

How LLMs Learn Low-, Medium-, and High-Effort Reasoning Modes

Optimizing your LLM in production

How To Build Your Own LLM Runtime From Scratch

If you have ever wanted to actually build an LLM inference runtime yourself — pack your own weights, own every barrier, capture your own CUDA graphs — this is what that journey looks like on an H100. A step-by-step tour of a small runtime called annotated-llm-runtime, and the three bugs that produced most of the annotations.

Welcome Gemma - Google’s new open LLM

Welcome Gemma 2 - Google’s new open LLM

Understanding and Coding the KV Cache in LLMs from Scratch

KV caches are one of the most critical techniques for efficient inference in LLMs in production.

Making LLMs lighter with AutoGPTQ and transformers

Judge Arena: Benchmarking LLMs as Evaluators