Expand description
MLX affine conventions over a safetensors inventory. This adapter does not interpret architecture-specific names or choose execution precision.
Functions§
- affine 🔒
- is_
component 🔒 - quantization 🔒
- tensors
- Combine MLX weight, scale, and bias entries into logical affine tensors. Per-layer quantization settings override global settings. Unknown encodings retain their metadata and data spans as opaque tensors; no weights are decoded.