oMLX creator Jun Kim joins Hugging Face to work full time on the Apple Silicon inference project
Hugging Face Blog·low signal
Hugging Face announced on September 22 that it is funding Jun Kim to maintain oMLX, an Apache 2.0 testbed for the MLX ecosystem built on mlx-lm and mlx-vlm, as a full-time role. The project stays Apache 2.0 and under Kim's lead. Hugging Face says it wants to streamline converting transformer model definitions into MLX implementations that multiple inference engines can use. Only Hugging Face's own post covers this so far.