feat: integrate deepgemm into EPMoE (#6821)

Co-authored-by: tianqilin.99 <tianqilin.99@bytedance.com> Co-authored-by: TianQiLin666666 <1834987979@qq.com> Co-authored-by: Cheng Wan <54331508+ch-wan@users.noreply.github.com>
2025-06-23 16:38:58 +08:00
parent 30ceccc74a
commit 506c4928f5
5 changed files with 336 additions and 3 deletions
--- a/python/sglang/test/test_block_fp8_ep.py
+++ b/python/sglang/test/test_block_fp8_ep.py
@@ -182,6 +182,7 @@ def ep_moe(
        end_expert_id,
        top_k,
        hidden_states.size(1),
+        0,
        BLOCK_SIZE=512,
    )
    return output