<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>AI Universe (English)</title><description>Guides and real-world cases of artificial intelligence, written by people who build with AI agents every day. If we haven’t tested something, we say so.</description><link>https://aiuniverse.blog/</link><language>en</language><item><title>Local models for coding vs Claude: how they actually perform and what they really cost</title><link>https://aiuniverse.blog/en/local-ai-coding-vs-claude/</link><guid isPermaLink="true">https://aiuniverse.blog/en/local-ai-coding-vs-claude/</guid><description>Qwen3.6, Devstral, gpt-oss and Muse vs Claude for coding: SWE-bench results, the hardware you need, and the math in euros for local versus the API.</description><pubDate>Sat, 17 Oct 2026 00:00:00 GMT</pubDate><category>Local AI and privacy</category><category>Local AI</category><category>Coding</category><category>Claude</category><category>Qwen3.6</category><category>gpt-oss</category><category>Cost</category></item><item><title>Local AI on a VPS without a GPU: which models work and how fast (we measured it)</title><link>https://aiuniverse.blog/en/local-ai-vps-no-gpu/</link><guid isPermaLink="true">https://aiuniverse.blog/en/local-ai-vps-no-gpu/</guid><description>We tested six models (Qwen3.5, Gemma 4, Ministral 3, Llama 3.2) with llama.cpp on a 3-vCPU VPS with no GPU: real speed, quality and what they&apos;re good for.</description><pubDate>Sat, 17 Oct 2026 00:00:00 GMT</pubDate><category>Local AI and privacy</category><category>Local AI</category><category>llama.cpp</category><category>VPS</category><category>No GPU</category><category>Qwen3.5</category><category>Gemma 4</category></item><item><title>The best local AI models in 2026 for your hardware</title><link>https://aiuniverse.blog/en/best-local-ai-models-hardware/</link><guid isPermaLink="true">https://aiuniverse.blog/en/best-local-ai-models-hardware/</guid><description>Which local AI model to pick for your RAM, GPU or Mac: Qwen3.5 and 3.6, Gemma 4, gpt-oss, Mistral and Muse, with memory needs, speed and license.</description><pubDate>Sat, 17 Oct 2026 00:00:00 GMT</pubDate><category>Local AI and privacy</category><category>Local AI</category><category>Open models</category><category>Qwen3.5</category><category>Gemma 4</category><category>gpt-oss</category><category>Hardware</category></item><item><title>Ollama, llama.cpp, or LM Studio: which to use for local AI in 2026</title><link>https://aiuniverse.blog/en/ollama-vs-llama-cpp-vs-lm-studio/</link><guid isPermaLink="true">https://aiuniverse.blog/en/ollama-vs-llama-cpp-vs-lm-studio/</guid><description>We compare Ollama, llama.cpp, and LM Studio, plus when to move to vLLM or Open WebUI: setup, API, licenses, performance, and what to pick for your use case.</description><pubDate>Sat, 17 Oct 2026 00:00:00 GMT</pubDate><category>Local AI and privacy</category><category>local AI</category><category>Ollama</category><category>llama.cpp</category><category>LM Studio</category><category>vLLM</category><category>Open WebUI</category></item><item><title>AI agents: real examples we use every day at a small business</title><link>https://aiuniverse.blog/en/ai-agents-real-examples/</link><guid isPermaLink="true">https://aiuniverse.blog/en/ai-agents-real-examples/</guid><description>Nine AI agents we use daily at GamzyHost: what they do, what they decide on their own, what they ask us, and which one you should build first.</description><pubDate>Fri, 16 Oct 2026 00:00:00 GMT</pubDate><category>AI for your business</category><category>AI agents</category><category>Claude Code</category><category>Skills</category><category>MCP</category><category>Real-world cases</category><category>Businesses</category></item><item><title>Claude Code: a complete guide, from zero to a serious workflow</title><link>https://aiuniverse.blog/en/claude-code-guide/</link><guid isPermaLink="true">https://aiuniverse.blog/en/claude-code-guide/</guid><description>Install Claude Code on Windows, Mac or Linux, run your first session and set up a serious workflow with permissions, CLAUDE.md, skills and subagents.</description><pubDate>Fri, 16 Oct 2026 00:00:00 GMT</pubDate><category>Coding with AI</category><category>Claude Code</category><category>agents</category><category>CLAUDE.md</category><category>MCP</category><category>skills</category><category>subagents</category><category>tutorial</category></item><item><title>CLAUDE.md: how to write it so Claude Code actually follows it</title><link>https://aiuniverse.blog/en/claude-md-guide-templates/</link><guid isPermaLink="true">https://aiuniverse.blog/en/claude-md-guide-templates/</guid><description>What to put in your CLAUDE.md, what to leave out, and how Claude Code loads it, with excerpts from the ones we use daily and two templates ready to copy.</description><pubDate>Fri, 16 Oct 2026 00:00:00 GMT</pubDate><category>Coding with AI</category><category>Claude Code</category><category>CLAUDE.md</category><category>Memory</category><category>Templates</category><category>Best practices</category></item><item><title>Claude pricing in 2026: plans, real limits, and which one is right for you</title><link>https://aiuniverse.blog/en/claude-pricing-plans-limits/</link><guid isPermaLink="true">https://aiuniverse.blog/en/claude-pricing-plans-limits/</guid><description>What Claude costs today: Free, Pro, Max, Team, Enterprise, and API, how the usage limits work, and which plan to pick based on how you&apos;ll use it.</description><pubDate>Fri, 16 Oct 2026 00:00:00 GMT</pubDate><category>AI assistants</category><category>Claude</category><category>Pricing</category><category>Claude Code</category><category>Claude API</category><category>Anthropic</category></item><item><title>What MCP (Model Context Protocol) is, with the servers we use every day</title><link>https://aiuniverse.blog/en/what-is-mcp/</link><guid isPermaLink="true">https://aiuniverse.blog/en/what-is-mcp/</guid><description>Learn what MCP is, how host, client and server fit together, how to add a server to Claude Code and which risks to watch, using our real servers.</description><pubDate>Fri, 16 Oct 2026 00:00:00 GMT</pubDate><category>AI agents and MCP</category><category>MCP</category><category>Model Context Protocol</category><category>Claude Code</category><category>Agents</category><category>Security</category></item><item><title>Claude skills: what they are and how to build your own, plus the 49 we use</title><link>https://aiuniverse.blog/en/claude-skills-create-your-own/</link><guid isPermaLink="true">https://aiuniverse.blog/en/claude-skills-create-your-own/</guid><description>Build a Claude skill that triggers when it should: the full SKILL.md, how to write the description, and what we learned from 49 skills in production.</description><pubDate>Fri, 16 Oct 2026 00:00:00 GMT</pubDate><category>AI agents and MCP</category><category>Claude Code</category><category>Agent Skills</category><category>SKILL.md</category><category>AI agents</category><category>Jarvis</category></item></channel></rss>