#LM Studio

2 stories found

Qwen3.8-27B open weights from Alibaba running on a laptop and a phone next to an Apache 2.0 licence badge

AIAug 16·3 min read

Qwen3.8-27B open weights under Apache 2.0, and the 17GB footnote

Alibaba released Qwen3.8-27B under Apache 2.0: 27 billion parameters, a 262K token context, and roughly 17GB to run it locally. That last number is the answer to the frontier-on-a-phone headline.

#Alibaba #Qwen #MediaTek

A laptop with integrated graphics running a local language model in LM Studio

ExplainersAug 16·6 min read

Run an Open Model on the Machine You Already Own, Integrated Graphics Included

What decides whether an open model loads is memory, not the badge on your GPU. Sizing the model, picking a quantisation, setting context deliberately, and knowing which jobs still belong to a hosted model.

#LM Studio #Ollama #llama.cpp