Thomas WangandXinyi Song f2bcdb0508 [AMD] Add unified kv attention support in dpsk-v4 (#27380) ...

Co-authored-by: Xinyi Song <86638975+RolaoDenthu@users.noreply.github.com>

2026-06-09 23:13:37 -07:00

..

attention

[AMD] Add unified kv attention support in dpsk-v4 (#27380)

2026-06-09 23:13:37 -07:00

deep_gemm_wrapper

Defer DeepGEMM PDL setup to worker init (#27671)

2026-06-09 13:52:30 -07:00

moe

[Refactor] Cuda Graph Runner/Backend Refactor (#23906)

2026-06-09 21:36:57 -07:00

quantization

[Refactor] Cuda Graph Runner/Backend Refactor (#23906)

2026-06-09 21:36:57 -07:00

rotary_embedding

[AMD] Optimize gpt-oss-120B performance (#27063)

2026-06-07 20:06:30 -07:00

triton_ops

[codex] Centralize Triton utility kernels (#26000)

2026-06-02 16:47:45 +08:00

utils

Bundle set_kv_buffer write targets into KVWriteLoc (loc + swa_loc) (#27695)

2026-06-09 23:09:51 -07:00

activation.py

[Refactor] Cuda Graph Runner/Backend Refactor (#23906)

2026-06-09 21:36:57 -07:00

amx_utils.py

[CPU] Add GPT-OSS model optimization for CPU (#16775)

2026-05-29 16:05:26 +08:00

clippable_linear.py

[New Model] Gemma 4 (#21952)

2026-04-06 20:24:44 -07:00

communicator_dsa_cp.py

DeepSeek V4: Support context parallelism with fused MoE (non-DeepEP) (#24947)

2026-06-01 14:25:43 -07:00

communicator.py

[Refactor] Cuda Graph Runner/Backend Refactor (#23906)

2026-06-09 21:36:57 -07:00

conv.py

[CPU] Add support for Qwen3-vl and Qwen3-omni (#12662)

2026-05-27 08:56:09 +08:00

deepseek_v4_rope.py

[AMD] Dsv4/pr2 compressor opt (#26208)

2026-05-25 23:54:40 -07:00

dp_attention.py

[Refactor] Rename NSA → DSA: user-facing aliases, file/class/import rename (#25821)

2026-05-20 00:18:04 -07:00

elementwise.py

[codex] Centralize Triton utility kernels (#26000)

2026-06-02 16:47:45 +08:00

flashinfer_comm_fusion.py

Remove FlashInfer GB transport workaround (#27453)

2026-06-09 16:48:49 -07:00

fused_qk_norm_rope_store.py

[AMD] Add unified kv attention support in dpsk-v4 (#27380)

2026-06-09 23:13:37 -07:00

fused_qk_norm.py

[AMD] Dsv4/pr2 compressor opt (#26208)

2026-05-25 23:54:40 -07:00

gemma4_fused_ops.py

Optimize Gemma4 H200 MoE and extend attention (#26588)

2026-06-06 14:14:25 +08:00

int4fp8_utils.py

[AMD][Quantization] Add int4fp8_moe online quantization on ROCm (#7392)

2026-01-14 01:44:40 -08:00

layernorm.py

[Refactor] Cuda Graph Runner/Backend Refactor (#23906)

2026-06-09 21:36:57 -07:00

linear.py

chore: add vLLM SPDX copyright headers to ported files (#25182)

2026-05-13 15:17:30 -07:00

logits_processor.py

Fix flashinfer autotune oom glm51 (#24195)

2026-06-02 23:28:57 -07:00

mhc_head.py

Optimize MHC pipeline: DeepGemm, fused norm, fused hc_head (#24775)

2026-05-10 19:03:37 +08:00

mhc.py

[DeepSeek-V4] Add mhc_fused_post_pre kernel (#25976)

2026-05-30 02:04:51 -07:00

model_parallel.py

[Refactor] Rename CustomOp -> MultiPlatformOp (#16175)

2025-12-31 01:16:32 +08:00

modelopt_utils.py

Enable native ModelOpt quantization support (1/3) (#7149)

2025-10-06 13:24:15 -07:00

multimodal.py

Fix weights_checker checksum for 0-dim tensors and multi-GPU (#26863)

2026-05-31 21:27:03 -07:00

n_gram_embedding.py

Optimize ngram decode id computation (#24757)

2026-06-02 17:37:34 +08:00

parameter.py

chore: add vLLM SPDX copyright headers to ported files (#25182)

2026-05-13 15:17:30 -07:00

pooler.py

[Score API] Add Multi-Item Scoring with pre-computed delimiter indices (#22544)

2026-04-20 22:50:40 -07:00

radix_attention.py

[Refactor] Cuda Graph Runner/Backend Refactor (#23906)

2026-06-09 21:36:57 -07:00

radix_linear_attention.py

[Refactor] Cuda Graph Runner/Backend Refactor (#23906)

2026-06-09 21:36:57 -07:00

rocm_linear_utils.py

[AMD] Fix DLPack Error in Aiter flydsl GEMM by Detaching MoE Gate Weight (#22262)

2026-04-07 23:42:10 -07:00

sampler.py

[NPU]adaptation to support deterministic inference (#21197)

2026-06-09 09:22:09 +08:00

sparse_pooler.py

Implement BGE-M3 Sparse Embeddings in SGLang (#10869)

2025-10-22 13:46:16 -07:00

torchao_utils.py

Lazy import torchao (#17626)

2026-01-22 22:04:51 -08:00

vocab_parallel_embedding.py

[Bug] Fix out-of-range token id crashing tp=1 VocabParallelEmbedding (#27482)

2026-06-06 23:00:17 -07:00