vllm/csrc/quantization/fp8 at d593cf28fa020dbae53dae19122aac8aeeeae0bc - vllm - Gitea: Git with a cup of macha paffe

karylab_agents/vllm

forked from Karylab-cklius/vllm

Files

T

History

elvischenvandGitHub dbeee3844c [Perf] Use NVIDIA hardware-accelerated instruction for float to fp8_e4m3 quantization (#24757 )

Signed-off-by: elvischenv <219235043+elvischenv@users.noreply.github.com>

2025-09-13 00:16:24 -07:00

..

[MISC] Remove unused variableds in C++ (#19609 )

2025-06-15 20:05:28 -07:00

[Perf] Use NVIDIA hardware-accelerated instruction for float to fp8_e4m3 quantization (#24757 )

2025-09-13 00:16:24 -07:00

common.cu

[Feature] Non-contiguous Support for FP8 Quantization (#21961 )

2025-08-05 02:36:43 -07:00

common.cuh

[Perf] Use NVIDIA hardware-accelerated instruction for float to fp8_e4m3 quantization (#24757 )

2025-09-13 00:16:24 -07:00

per_token_group_quant.cu

[Perf] Using __nv_fp8_e4m3 instead of c10::e4m3 for per_token_group_quant (#21867 )

2025-07-29 21:50:46 -06:00