Skip to content
BranchesTags

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
triton_3_4_0
triton_3_4_0
 
 
E=257,N=512,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8.json
E=257,N=512,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8.json
 
 
N=1024,K=7168,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
N=1024,K=7168,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
 
 
N=2112,K=7168,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
N=2112,K=7168,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
 
 
N=24576,K=7168,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
N=24576,K=7168,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
 
 
N=3072,K=1536,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
N=3072,K=1536,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
 
 
N=3072,K=7168,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
N=3072,K=7168,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
 
 
N=32768,K=512,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
N=32768,K=512,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
 
 
N=4096,K=512,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
N=4096,K=512,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
 
 
N=4608,K=7168,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
N=4608,K=7168,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
 
 
N=512,K=7168,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
N=512,K=7168,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
 
 
N=576,K=7168,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
N=576,K=7168,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
 
 
N=6144,K=1536,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
N=6144,K=1536,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
 
 
N=7168,K=16384,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
N=7168,K=16384,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
 
 
N=7168,K=18432,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
N=7168,K=18432,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
 
 
N=7168,K=2048,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
N=7168,K=2048,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
 
 
N=7168,K=2304,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
N=7168,K=2304,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
 
 
N=7168,K=256,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
N=7168,K=256,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
 
 
N=7168,K=4096,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
N=7168,K=4096,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
 
 
N=7168,K=4608,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
N=7168,K=4608,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
 
 
N=7168,K=512,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
N=7168,K=512,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
 
 
N=8192,K=512,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
N=8192,K=512,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
 
 
N=9216,K=7168,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
N=9216,K=7168,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8,block_shape=[128, 128].json
 
 
README.md
README.md
 
 

Repository files navigation

sglang-quantization-configs

NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition/NVIDIA_RTX_PRO_6000_Blackwell_Workstation_Edition configs

cd python/sglang/srt/layers/quantization/configs

cd python/sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0

About

No description, website, or topics provided.

Resources

Readme
Activity

Stars

1 star

Watchers

0 watching

Forks

1 fork
Report repository

Releases

Packages

Contributors