ref(upstream): add Deep-Spark xllm + vllm MoE/GDN reference code

Browse Source

Sources (Apache 2.0, cloned 2026-08-09):
- Deep-Spark/xllm: Iluvatar's official C++ inference engine
- Deep-Spark/vllm: Iluvatar's vllm fork

Key files for our EX Engine development:

MoE topk_softmax (fixes 2304 calls/token PyTorch fallback):
- xllm/kernels/cuda/moe/moe_topk_softmax_kernels.cuh
  CUB-based fused softmax+topk, power-of-2 expert count optimized
  For 64 experts: topk_gating_softmax<T,VPT=2,64,WARPS=4,BYTES=4>
- xllm/kernels/ilu/ixformer.h
  Official ixformer C++ API: topk_softmax(), paged_attention(), etc.
- xllm/kernels/ilu/fused_moe.cpp
  How xllm calls ixformer::infer::topk_softmax()
- ds_vllm/csrc/moe/topk_softmax_kernels.cu
  vllm-native topk_softmax (TensorRT-LLM derived, 874 lines)

GatedDeltaNet (fixes NaN in 4 GDN layers):
- xllm/layers/npu_torch/qwen3_gated_delta_net_base.cpp
  fp32 state accumulation, proper recurrent update

Complete FusedMoE pipeline reference:
- xllm/layers/ilu/fused_moe.cpp
  gate -> topk -> expand -> gemm1 -> act -> gemm2 -> combine

...

This commit is contained in:

EX Engine

2026-08-10 02:48:17 +00:00

parent b4e055e9a9

commit ea82b00e54

33 changed files with 13042 additions and 0 deletions

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Download Patch File Download Diff File Expand all files Collapse all files

3998

upstream_ref/ds_vllm/vllm/_custom_ops.py Normal file

File diff suppressed because it is too large Load Diff