Skip to content

Sebastian Raschka personal blog

LLM architecture comparison

by Sebastian Raschka, PhD magazine.sebastianraschka.com published 2025-07-19

From DeepSeek-V3 to Kimi K2: A Look At Modern LLM Architecture Design

Read on magazine.sebastianraschka.com → Discuss on HN (24) Archived copy

Points: 418
HN comments: 24
Submissions: 1
Cited by: 0 commenters

AI & Machine Learning Software Engineering 2025

On Hacker News

Submitted	Title on HN	Points	Comments
Jul 20, 2025	LLM architecture comparison by mdp2021	418	24

More from Sebastian Raschka

All 17 posts →

Building LLMs from the Ground Up: A 3-Hour Coding Workshop

2024970 points136 comments AI & Machine Learning details

Sebastian Raschka's from-scratch video course on building reasoning models

2026856 points116 comments AI & Machine Learning details

GPT-6 Astra, looped transformers, and hidden reasoning

2026521 points162 comments AI & Machine Learning details

GPT-OSS vs. Qwen3 and a detailed look how things evolved since GPT-2

2025490 points97 comments AI & Machine Learning details

Understanding Reasoning LLMs

2025473 points183 comments AI & Machine Learning details

Practical Tips for Finetuning LLMs Using LoRA (Low-Rank Adaptation)

2023342 points27 comments AI & Machine Learning details

Related posts from the same era

Statement on US government directive to suspend access to Fable 5 and Mythos 5

Anthropic20263,158 points2,314 commentscited by 11 details

GPT-4o

OpenAI Blog20243,138 points2,366 commentscited by 16 details

Astra can read and write weights to improve Qwen

Rikiya Takehi20262,950 points895 comments details

HackerTube: YouTube built from videos recommended by HN

karpathy20262,882 points640 comments details

Claude Fable 5

Anthropic20262,626 points2,159 commentscited by 10 details

Claude Code is steganographically marking requests

Thereallo20262,445 points750 commentscited by 3 details