pub fn weight_rows(gpu: &dyn GpuBackend, weight: &QuantizedWeight) -> u32Expand description
2026-09-28: The widest row count the W4A4 path serves for weight on gpu under the
published tier (WeightQuantTier::w4a4_rows); 0 means W4A16.
pub fn weight_rows(gpu: &dyn GpuBackend, weight: &QuantizedWeight) -> u322026-09-28: The widest row count the W4A4 path serves for weight on gpu under the
published tier (WeightQuantTier::w4a4_rows); 0 means W4A16.