I've done some things

I try every new small model that comes along.

The best way to buy AI inference is usually a subscription.

I don’t like the verbosity of AI models and I don’t generally like their prose.

I like Claude.

Updated July 6th with Reasonix, July 17th with Kimi Code, July 22nd with Pool, July 23rd with Qwen Coder and MiMo Code.

Updated July 21 with GPT 5.6 Sol.

In discussions about my recent pile of benchmarks of LLMs ability to crack hard security bugs (Mythos-discovered bugs), such as here and here, more than a few folks have had questions about how to run models locally.

I recently added Ornith 1.0 to the Will It Mythos benchmark.

In a previous post, I tested a bunch of models on the problem of finding bugs in the wild originally reported by Mythos.

I grew up on the Commodore 64, so sometimes I like to tinker with those old 8-bit machines, sometimes the real thing, usually an emulator.

Open models seem like a democratizing force, right?

Qwen 3.6 punches well above its weight in hunting hard security bugs, as I noted in a previous post.

Updated on June 17th, 21st, 22nd, 24th, 26th, and 27th, and July 3rd.

IBM recently released a big pile of new open weight foundation models, Granite version 4.1, and there’s a few interesting things going on with these models.

Agents can write a lot of code, fast, and with the best models, most of it even works.

We recently had an issue on a customer’s system where DNS wasn’t responding.

I was recently talking with our awesome UI/UX designer, Ilia, about a problem he was having with Perl.