[ Video ] How To Set Up Local AI on a 16GB MacBook
Setting up local AI on a 16GB MacBook involves using optimized binaries for the Metal architecture. Start with MLX-serve. Optionally OpenWebUI
What Makes an AI Model a “Really Good” AI Model?
At the end of the day, all AI models are the same: statistical approximation machines. A really good AI model is just in the question you [...]
[ Video ] Running Xiaomi MiMo V2.6 (Qwen 9B distill) on my 16 GB MacBookPro
This week I'm testing Xiaomi MiMo V2.6, Qwen 9b distill, on a 16GB M1 MacBookPro. It works good enough for text generation and light coding
[Video] Running Google Gemma 4 E4B on my 16GB MacBookPro
This week I'm testing Google Gemma 4 E4B on a 16GB M1 MacBookPro. It works surprisingly well for text generation and even light coding.
[Video] I Tested Grok 4.7 With One Simple Job: Turn My vi Task Files Into a Bash Agenda
Grok 4.7 was launched yesterday and I had a chance to test it. It has better prompt adherence, it's more thorough, but also token greedy.
[ Video ] A System One AI Doesn’t Talk, It Only Decides – And It’s Perfect for Assess Decide Do
A System One model, like JEV, doesn't output text, it only decides, and that's why is perfect for my Assess Decide Do framework.
[Video] Running Qwen 3.8 27B Ternary Bonsai 2 on my 16GB Mac:
This is a brief test of a heavily quantized version of Qwen 3.8 27B, fitting in 7GB on disk. Impressive, but not 98% of the original [...]
The AI Agents Took Over My Blog Traffic. I Made Them Pay For That…
95% of my blog is coming from agents now. They didn't ask for permission, those bots, so I decided to make the AI agents pay.
What Open Weights Models Can You Realistically Run on a Medium Laptop (September 2026)?
Open source AI is here to stay, so let's see what open weights models you can realistically run on a medium laptop as of September 2026.
Anthropic Wants to Slow Down AI, Ban Open Source, and Still Aims an IPO at $2T Valuation. Really?
In a rather disturbing essay, Dario Amodei, Anthropic's CEO, urges to slow down AI dissemination (not training) and ban open source AI.
An App-pocalypse Is Coming! But Does It, Really?…
It's very easy to vibe code an entire SaaS these days. But the cost of building going down doesn't mean ALL the apps will disappear.
The Complete Vibe Coding Toolkit for Lazy, Creative People
Vibe coders are not coders. And that might be their unfair advantage. For these lazy, creative people I built a vibe coding toolkit.
You Can Now Share Your Grok Bots (And Soon You Could Sell Them)
You can now share your Grok bots. I had early access to the feature and I published my own before they went live. I explain how [...]
Grok Bot is the “iPhone Moment” for AI Agents – Quick Review
Grok Bot quick review: it looks simple, but at the same time useful, just like the first iPhone. Agents talking to each other feels magical.
How I Use Grok Build to Run Free Models Alongside Grok-4.6
Use Grok Build with local and OpenRouter free models alongside Grok-4.6. Copy-paste config.toml for Gemma, Qwen, and GLM — same harness, switch with /model.
[ Video ] How To Choose an Open Weights Model – Episode 4
In the final episode I talk about open weight model harnesses, or how to actually talk to your local AI. From llama-cpp chat to Pi, or [...]
[ Video ] How To Choose an Open Weights Model – Episode 3
In this episode I talk about running an open weights model locally. Two main hardware lines: Apple Silicon and NVIDIA, both good.
[ Video ] How To Choose an Open Weights Model – Episode 2
Episode 2 of the mini series about how to choose an open weights model, this time how to choose an open weights model provider.
[ Video ] How To Choose an Open Weights Model – Episode 1
How to choose and open weights model: the difference between slow, dumb and laggy - and state of the art inference, like Claude or Grok.
How AI Watermarking Works? Can It Be Removed?
AI watermarking is not baked into model weights, it works at the generation level, by sampling different tokens, paired with a private key.