Logo
working models | blog
Search
login
workingmodels.ai ↗
videos ↗
Subscribe
Logo

gemma4

gemma4

GGUF vs MLX on Apple Silicon

Jun 18, 2026

•

3 min read

GGUF vs MLX on Apple Silicon

Everyone says the same thing: MLX is faster on Apple Silicon, and it makes sense, MLX was built for Apple's chips. I wanted to see if that holds up, so I tested it. I took Gemma 4 26B in both formats, GGUF and MLX, and ran the same tests on each, on my M3 Ultra.

Bartlomiej Slodyczka
Bartlomiej Slodyczka

Working Models

Weekly lab notes from my AI builds: what I tried, what actually happened. Configs and numbers kept in.

© 2026 Working Models.
beehiivPowered by beehiiv