A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 15, 01:00 UTC
[Doc] Remove duplicate docstring (#21012)
Signed-off-by: yewentao256 <zhyanwentao@126.com>
Wentao Ye committed
Jul 15, 2025 at 23:09 UTC
76ddeff2931d1a5bc4192815c6ed778541e9f59e
1 file changed
-2
vllm/model_executor/layers/quantization/utils/fp8_utils.py
-2
@@ -378,8 +378,6 @@ def per_token_group_quant_fp8(
378
is supported for now.
379
column_major_scales: Outputs scales in column major.
380
out_q: Optional output tensor. If not provided, function will create.
381
- tuple[torch.Tensor, torch.Tensor]: The quantized tensor and the
382
- scaling factor for quantization.
381
Returns:
382
tuple[torch.Tensor, torch.Tensor]: The quantized tensor and the
383
scaling factor.