FastDeploy/fastdeploy/model_executor/layers/moe at 4e5e36ec9c8d15a7c505b93f77c7a1d48f16e87a - FastDeploy - 子说镜像小站

apps/FastDeploy

mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-12-24 13:28:13 +08:00

Files

History

bukejiyu 71781b56e1 RL fix (#5505 )

2025-12-11 19:25:24 +08:00

..

__init__.py

support w4afp8 EP inference (#3044 )

2025-08-25 11:27:45 +08:00

ep.py

[Feature] support Two batch overlap, mainly used in Prefill (#5078 )

2025-12-05 14:58:50 +08:00

fused_moe_backend_base.py

RL fix (#5505 )

2025-12-11 19:25:24 +08:00

fused_moe_cutlass_backend.py

[New][RL] Support Rollout Routing Replay (#5405 ) (#5408 )

2025-12-08 10:00:35 +08:00

fused_moe_deepgemm_backend.py

[New][RL] Support Rollout Routing Replay (#5405 ) (#5408 )

2025-12-08 10:00:35 +08:00

fused_moe_marlin_backend.py

[New][RL] Support Rollout Routing Replay (#5405 ) (#5408 )

2025-12-08 10:00:35 +08:00

fused_moe_triton_backend.py

[New][RL] Support Rollout Routing Replay (#5405 ) (#5408 )

2025-12-08 10:00:35 +08:00

fused_moe_wint2_backend.py

[New][RL] Support Rollout Routing Replay (#5405 ) (#5408 )

2025-12-08 10:00:35 +08:00

moe.py

[New][RL] Support Rollout Routing Replay (#5405 ) (#5408 )

2025-12-08 10:00:35 +08:00

routing_indices_cache.py

[New][RL] Support Rollout Routing Replay (#5405 ) (#5408 )

2025-12-08 10:00:35 +08:00

triton_moe_kernels.py

[OPs] MoE support wfp8afp8(channelwise) and improve per_token_quant_fp8 (#4238 )

2025-09-24 16:39:51 +08:00