Kimi k3 vs. Claude Opus 3.5: Can Open Weights keep up?

Claude Opus 3.5: Same Prompt – Big Difference? (Low vs. High vs. Max)

Claude Opus 5 is here – and I'm thrilled

Vibe Coding done right: How to make your AI prototype scalable

Can Kimi K3 code an entire game? (Practical Test)

n8n was too much for me: My local AI automation

Save Your Money: GPT 5.6 Luna, Terra, or Sol?

Local AI directly in your browser – no LM Studio or Ollama required!

AI Without Cloud or Subscriptions: LM Studio Tutorial for Local AI

Local LLM (Ornith 9B) thinking too long? Here is the solution

Fable 5 vs. Sonnet 5: Expensive AI just for planning?

Ornith 1.0 surprises me... but it's also annoying

Large models via Ollama Cloud: How far can you get with the free tier?

I wanted to use GitHub Copilot for free - here's what happened

Agent Coding in VS Code: Which Extension Does It Best?

Local Coding Agents: Do Workflow Tricks Save Small Models?

Hot or cold? The best LLM temperature for AI coding

Fable 5 vs. Opus 4.8: Is the expensive AI coding model really worth it?

Gemma 2 QAT tested: Is 4-bit better than 8-bit??

Gemma 4 12B Tested: The Local AI Sweet Spot for Agentic Coding?

Google Antigravity 2.0: Now like Codex & Claude's Code?

I'm having 3 AI plans implemented by the local Gemma 4.

Opus 4.8 builds my calculator test. Is it perfect?

4-Bit, 6-Bit, 8-Bit: Which Quantization Is Best for Gemma?

My AI Workspace: How I organize all my projects with AI

LM Studio, Ollama, or direct? It depends…

Using Codex with a local LLM – How to get the connection working

Understanding Local LLM Agents: My Test Lab for Tool Calls & File Edits

Chrome Prompt API released: Local AI tested for websites

TanStack Supply Chain Attack: How to protect yourself from dangerous npm packages

Gemma 4 is thinking itself to death: Agentic Coding with and without Thinking

Can my local AI model compete with free OpenCode models?

Small models, big effort? OpenCode put to the test.

Qwen 3.6 Q2-Bit im Agent Coding Test

Local agentic coding: How small can the model be?

Codex + Browser Use: GPT-5.5 builds and tests a React app itself

OpenCode tested in the terminal: Qwen 3.6 or Gemma 4?

Qwen 3.6 tested locally: Better than Gemma 4 for agentic coding?

Claude Design: Prompt + sketch → finished prototype (with code!)

Claude Opus 4.7 is here – better than 4.6 & GPT 5.4?

AI fatigue in programming – does AI coding really make us more productive?

Caveman skill for agentic coding: What are the benefits?

Gemma 4 in VS Code (Continue): Agentic Coding with a Small Model

Gemma 4 locally in the agent workflow: Is it suitable for programming?

Google Gemma 4: finally usable for coding?!

Axios Hack: Making npm Installs More Secure

Wie ich mit Claude von unterwegs programmiere ohne Laptop

Anti-Vibe Coding: Does this make AI coding fun again?

AI as a QA tester? Claude Computer Use on macOS – first test

MacBook Neo for web development. Is it really worth it?

Affordable AI models for coding: Is it really worth it?

Cursor without limits? Here's how you can still use it with other (free) models.

Claude Code: Cheaper, but still good? My practical experience

The biggest prompting hack in Claude Code: Skills

Get more VRAM on your Mac with this trick: How to load larger AI models

The biggest mistake in AI coding: How to avoid it

Der OpenAI API Turbo: Responses jetzt via WebSockets

Claude Code without a terminal! Desktop app put to the test 🚀

Replacing Claude Cowork? My Codex setup put to the test.

Use Claude Code for FREE (locally with Ollama / LM Studio) – no Anthropic subscription required.