The guide to local image creation with Z-image

The longer guide to installing Crush with Deepseek

This is a great illustration of the potential LLMs have to quickly analyze data

GPT-5, the world best model just 1 year ago, is today inferior to Qwen3.6 27B that you can run on your desktop

AI Didn’t Make Programming Easier. It Just Made It Differently Difficult

Baidu open-sourced a peanut-sized OCR that parses entire 100-page PDFs in one shot

"Useful" is not sufficient

RWKV is an RNN with great LLM performance and parallelizable like a Transformer.

An open-source 26M parameter function-calling model that runs at 6000 tok/s prefill and 1200 tok/s decode on consumer devices.

Burgerland media is already starting a fear mongering campaign against GLM

Run GLM-5.2 (744B MoE) on a 25GB-RAM consumer machine

Setting Deepseek on translating a Japanese PS2 game to English [experimental]

How Deepseek reorganized my ROM collection (the good and bad)

Working With AI: A Concrete Example

Qwen 3.6 27B is the sweet spot for local development

Deepseek announces full V4 release; peak hour pricing increase

MAX is an engine like Llama.cpp that's entirely self contained with the same codebase and kernels targeting different hardware without needing Cuda or MLX kernels.

Ornith-1.0: Self-Scaffolding LLMs for Agentic Coding

GLM-5.2 is the step change for open agents

VibeThinker is a 3B param model that beats Opus 4.5 on reasoning with novel SFT+GRPO