KnownWeightPrecision enum
The precision a model's weights are served at.
Fields
| BF16 | 16-bit brain floating point. |
| FP16 | 16-bit floating point. |
| FP32 | 32-bit floating point. |
| FP4 | 4-bit floating point. |
| FP8 | 8-bit floating point. |
| INT4 | 4-bit integer. |
| INT8 | 8-bit integer. |