Topic
Local and Private AI
Running models on your own hardware: Ollama, LM Studio, vLLM, llama.cpp, GPUs and Macs for LLMs.5 posts
AI ModelsMuse Spark Open: Meta's Open-Weight Model, Licence and How to Run ItMuse Spark Open is Meta's 816B open-weight MoE model. What it is, how it relates to Muse Wonder, the licence terms, hardware needs and how to run or access it. · 5 min read
Local and Private AIBest Local LLM for Coding in 2026 (Picked by Your Hardware)The best local coding models for 16, 24-32 and 64 GB+ machines, with sizes, licenses and vendor benchmarks, plus how to plug them into Claude Code or your IDE. · 7 min read
Local and Private AILocal LLM Hardware Guide: How Much VRAM and RAM You Really NeedWork out the memory any local LLM needs: weights by quantization, KV cache for context, a sizing table from 3B to 120B, and how GPUs and Macs compare. · 8 min read
Local and Private AIOllama vs LM Studio: Which Local LLM Tool Should You Use?Ollama is the open-source, CLI-first runner for developers; LM Studio is the polished desktop app. Compare engines, APIs, platforms and licensing, and pick one. · 7 min read
Local and Private AIHow to Run an LLM Locally in 2026: Tools, Hardware and ModelsRun an LLM on your own computer: compare Ollama, LM Studio, llama.cpp, vLLM and Jan, size memory for any model, set up Ollama step by step and pick a model. · 9 min read