Browse Tag

Large Language Models

3 Articles

Qwen 3.8 27B: Vision, 262K Context & Apache 2.0

By Sutopo

9 Min Read

Alibaba's Qwen 3.8 27B combines native vision, a 262K-token context window, and Apache 2.0 licensing in one commercially deployable open-weight model.

Run a 27B Model Locally in LM Studio

By Sutopo

9 Min Read

A practical guide to running a 27B parameter AI model locally with LM Studio, covering hardware requirements, GGUF vs MLX format selection, and API server setup.

LLM Numbers: Parameter Count, Training Data Size, and Model Performance

LLM Parameters and Training Data: A Beginner’s Guide

By Sutopo

8 Min Read

Unlocking the secrets of large language models (LLMs): Understand LLM parameters and training data size. Learn how these numbers impact performance, from basic responses to complex tasks.