Mixed-precision MLX quantization of darkc0de/XORTRON.CriminalComputing.2026.27B.Instruct.NEXT, quantized with MLX Smart Quantize (MSQ) — my own sensitivity-based mixed-precision quantization method for Apple Silicon. It measures per-layer NMSE and assigns optimal bit widths automatically, combining architecture knowledge with measured data.

Details

Type: Vision (VLM)
Average: 4.5 bits per weight
Method: MLX Smart Quantize (MSQ)

mlx-community/XORTRON.CriminalComputing.2026.27B.Instruct.NEXT-4.5bit-msq

XORTRON.CriminalComputing.2026.27B.Instruct.NEXT — MLX 4.5 BPW

Details

mlx-community/
XORTRON.CriminalComputing.2026.27B.Instruct.NEXT-4.5bit-msq