Would love to run PersonaPlex on Apple Silicon Macs via MLX. The 7B model size is well within M-series memory limits and other speech models (Whisper, Moshi) already have MLX ports that work great.
Is this on the roadmap or would a community PR be welcome?
Would love to run PersonaPlex on Apple Silicon Macs via MLX. The 7B model size is well within M-series memory limits and other speech models (Whisper, Moshi) already have MLX ports that work great.
Is this on the roadmap or would a community PR be welcome?