public GpuQuantizedMatrix(path as Path, name as String, device as Device = Device.gpu)
Loads one Q8 tensor without retaining the open file.
| Parameter | Description |
|---|---|
path | Klyn Q8 v1 shard. |
name | Tensor name. |
device | GPU selector; auto chooses the first compatible GPU only. |
matrix = GpuQuantizedMatrix(Path("model.safetensors"), "lm_head.weight", Device.gpu[0])