Wenbo Yang 75b656488a Support serving DeepSeek-R1-Channel-INT8 with 32 L40S. (#4418)

2025-03-17 00:03:43 -07:00

..

ep_moe

cleanup deps 1/n (#4400)

2025-03-14 00:00:33 -07:00

fused_moe_triton

Support serving DeepSeek-R1-Channel-INT8 with 32 L40S. (#4418)

2025-03-17 00:03:43 -07:00

fused_moe_native.py

Support penalty in overlap mode; return logprob with chunked prefill; improve benchmark scripts (#3988)

2025-03-03 00:12:04 -08:00

router.py

Add some fused elementwise kernels for grok-1 (#4398)

2025-03-13 13:39:10 -07:00

topk.py

use topk_softmax with sgl-kernel (#4439)

2025-03-14 15:59:06 -07:00