lzy 690bcb8e50 [Optimization] 1.fix tp+ep moe_forward; 2.set max_prefill_batch=env.MAX_PREFILL_NUM (#5315)

2025-12-03 13:33:15 +08:00

..

__init__.py

support w4afp8 EP inference (#3044)

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

ep.py

[Quantization] Support w4afp8 MoE dynamic quantization (#5282)

2025-12-02 18:56:16 +08:00

fused_moe_backend_base.py

[Intel HPU] change MoE weights and scales from list to tensor and add… (#5289)

2025-11-28 19:17:05 +08:00

fused_moe_cutlass_backend.py

[Quantization] Support w4afp8 MoE dynamic quantization (#5282)

2025-12-02 18:56:16 +08:00

fused_moe_deepgemm_backend.py

[Optimization] Refine row parallel bias and nranks and moe all_reduce (#5247)

2025-11-26 05:09:09 -08:00

fused_moe_marlin_backend.py

[Optimization] Refine row parallel bias and nranks and moe all_reduce (#5247)

2025-11-26 05:09:09 -08:00

fused_moe_triton_backend.py

[Optimization] Refine row parallel bias and nranks and moe all_reduce (#5247)

2025-11-26 05:09:09 -08:00

fused_moe_wint2_backend.py

[Optimization] Refine row parallel bias and nranks and moe all_reduce (#5247)

2025-11-26 05:09:09 -08:00

moe.py

[Optimization] 1.fix tp+ep moe_forward; 2.set max_prefill_batch=env.MAX_PREFILL_NUM (#5315)

2025-12-03 13:33:15 +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