vllm-project/vllm

Pull Requests Commits

[Misc] Copy HF_TOKEN env var to Ray workers (#21406)

ruisearch42 committed 140 days ago

Verified 35bc8bd5

[Model] Add Qwen3CoderToolParser (#21396)

ranpox committed 141 days ago

Verified 4594fc3b

Fix Flashinfer Allreduce+Norm enable disable calculation based on `fi_allreduce_fusion_max_token_num` (#21325)

xinli-sw committed 141 days ago

Verified ae268b63

[CI/Build] Fix test failure due to updated model repo (#21375)

DarkLight1337 committed 141 days ago

Verified 35366ae5

[Bugfix] Decode Tokenized IDs to Strings for `hf_processor` in `llm.chat()` with `model_impl=transformers` (#21353)

ariG23498 committed 141 days ago

Verified 2226d5bd

Add tokenization_kwargs to encode for embedding model truncation (#21033)

Receiling committed 141 days ago

Verified 44554a00

Revert "[Refactor] Fix Compile Warning #1444-D (#21208)" (#21384)

yewentao256 committed 141 days ago

Verified 226b452a

[feat] Enable mm caching for transformers backend (#21358)

zucchini-nlp committed 141 days ago

Verified f38ee34a

Adds parallel model weight loading for runai_streamer (#21330)

bbartels committed 141 days ago

Verified b194557a

[Perf] Cuda Kernel for Per Token Group Quant (#21083)

yewentao256 committed 141 days ago

Verified 774d0c01

[feat]: add SM100 support for cutlass FP8 groupGEMM (#20447)

djmmoss committed 141 days ago

Verified 2c8db17c

[perf] Add fused MLA QKV + strided layernorm (#21116)

mickaelseznec committed 141 days ago

Verified 4fb56914

[Misc] unify variable for LLM instance v2 (#21356)

andyxning committed 141 days ago

Verified 0df4d9b0

[Core] Introduce popleft_n and append_n in FreeKVCacheBlockQueue to further optimize block_pool (#21222)

Jialin committed 141 days ago

Verified ed250545

[benchmark] Port benchmark request sent optimization to benchmark_serving (#21209)

Jialin committed 141 days ago

Verified 10904e6d

[Core] Optimize update checks in LogitsProcessor (#21245)

Jialin committed 141 days ago

Verified a3223766

[Misc] Remove deprecated args in v0.10 (#21349)

kebe7jun committed 141 days ago

Verified bc8a8ce5

[Bugfix] Fix eviction cached blocked logic (#21357)

simon-mo committed 141 days ago

Verified 32142b3c

Add arcee model (#21296)

alyosha-swamy committed 141 days ago

Verified 82b8027b

[Feature][eplb] add verify ep or tp or dp (#21102)

lengrongfu committed 141 days ago

Verified 3779eb8c

Update fp4 quantize API (#21327)

wenscarl committed 141 days ago

Verified 9e23ad96

[Bug] DeepGemm: Fix Cuda Init Error (#21312)

yewentao256 committed 141 days ago

Verified e69a92a1

[Misc] DeepEPHighThroughtput - Enable Inductor pass (#21311)

varun-sundar-rabindranath committed 141 days ago

Verified 8425f785

Fix kv_cache_dtype handling for out-of-tree HPU plugin (#21302)

kzawora-intel committed 141 days ago

Verified c17231e8

[Refactor] Fix Compile Warning #1444-D (#21208)

yewentao256 committed 141 days ago

Verified 6e5b5ca5

[V1] [Hybrid] Add new test to verify that hybrid views into KVCacheTensor are compatible (#21300)

tdoublep committed 141 days ago

Verified 488d8a98

[Core] Minimize number of dict lookup in _maybe_evict_cached_block (#21281)

Jialin committed 141 days ago

Verified af376ca1

Revert "[Performance] Performance improvements in non-blockwise fp8 CUTLASS MoE (#20762) (#21334)

minosfuture committed 141 days ago

Verified e7b20426

[Intel GPU] Ray Compiled Graph avoid NCCL for Intel GPU (#21338)

ratnampa committed 141 days ago

Verified 90f1e554

[Doc] Fix CPU doc format (#21316)

bigPYJ1151 committed 141 days ago

Verified 5e70dcd6

Newer Older