EX Engine d00daa62f6 feat(MoE): full ixformer pipeline — topk → gen_idx → expand → group_gemm → silu → combine ...

Port complete MoE pipeline from upstream xllm/layers/ilu/fused_moe.cpp.
All 6 ixformer::infer functions now exposed via ix_moe_bridge.cpp:

  1. topk_softmax          — fused routing (was: 3 PyTorch ops)
  2. moe_compute_token_index_api — build permutation maps
  3. moe_expand_input      — gather tokens by expert
  4. moe_w16a16_group_gemm — batched expert GEMM (was: Python for-loop)
  5. silu_and_mul           — fused activation
  6. moe_output_reduce_sum — weighted scatter-add

qwen3_5.py dispatch order:
  1. Try ix_fused_moe_forward (full C++ pipeline, 7 kernel launches)
  2. Try ix_topk_softmax only + PyTorch GEMM
  3. Pure PyTorch fallback (torch.softmax + torch.topk + for-loop)

ix_bridge.py exposes both individual ops and fused_moe_forward().
No upstream code copied — only forward-declarations of ixformer C++ API
that the base image SDK already contains.

2026-08-10 03:06:14 +00:00

..

csrc

feat(MoE): full ixformer pipeline — topk → gen_idx → expand → group_gemm → silu → combine

2026-08-10 03:06:14 +00:00

include

feat(EX): Algorithm Factor Replacement Engine — dlopen-based CUDA kernel dispatch

2026-08-10 02:25:23 +00:00

python

fix: fail-fast on ix_bridge failure + probe script for real machine

2026-08-10 03:04:50 +00:00

build.sh

refactor(EX): upstream-aligned kernels + FlashQLA GDN backend

2026-08-10 02:55:58 +00:00