Still need more time with it! Fell below gemma4 and gpt-oss from my initial bench run with it but need to try some other runtimes

Or macbook pro, or a 7 year old GPU even (if you offload)! Extremely versatile for it's size

gpt-oss:20b is still maybe the BEST all around small model (as far as quality, size, speed, breadth) gemma4:26b arguably better but only time will tell. gpt-oss has been SOLID for like a year at this point though, it has LEGS!

It's all about the OPEN frontier. Closed source lockin is just a rugpull waiting to happen

Lesson 5 of the PlebDevs Adversarial AI course is now live! https://plebdevs.com

My hermes agent (running gemma4:26b locally btw) used the new manim-video skill to make a video diagram describing how hermes works - not bad!

gpt-oss 🤝 gemma4 Open source American AI excellence 🇺🇸

Ollama powered by MLX on Apple now. This is huge for local models...