A visual guide to attention variants in modern llms Sebastian Raschka · Ahead of AI ·Mar 22, 2026 · 25 min read · AI & Tech
A dream of spring for open-weight llms: 10 architectures from jan-feb 2026 Sebastian Raschka · Ahead of AI ·Feb 25, 2026 · 25 min read · AI & Tech
From DeepSeek v3 to v3.2: Architecture, sparse attention, and rl updates Sebastian Raschka · Ahead of AI ·Dec 3, 2025 · 29 min read · AI & Tech
Understanding the 4 main approaches to LLM evaluation Sebastian Raschka · Ahead of AI ·Oct 5, 2025 · 32 min read · AI & Tech
From gpt-2 to gpt-oss: Analyzing the architectural advances Sebastian Raschka · Ahead of AI ·Aug 9, 2025 · 28 min read · AI & Tech
The big LLM architecture comparison Sebastian Raschka · Ahead of AI ·Jul 19, 2025 · 55 min read · AI & Tech
Understanding and coding the kv cache in llms from scratch Sebastian Raschka · Ahead of AI ·Jun 17, 2025 · 15 min read · AI & Tech
The state of reinforcement learning for LLM reasoning Sebastian Raschka · Ahead of AI ·Apr 19, 2025 · 38 min read · AI & Tech
The state of LLM reasoning model inference Sebastian Raschka · Ahead of AI ·Mar 8, 2025 · 23 min read · AI & Tech
Noteworthy AI research papers of 2024 Sebastian Raschka · Ahead of AI ·Jan 15, 2025 · 28 min read · AI & Tech
Noteworthy AI research papers of 2024 Sebastian Raschka · Ahead of AI ·Dec 31, 2024 · 17 min read · AI & Tech
LLM research papers: The 2024 list Sebastian Raschka · Ahead of AI ·Dec 8, 2024 · 30 min read · AI & Tech
New LLM pre-training and post-training paradigms Sebastian Raschka · Ahead of AI ·Aug 17, 2024 · 23 min read · AI & Tech
