vllm-project/vllm

Pull Requests Commits

[Doc] Update description disable_any_whitespace (#29784)

FredericOdermatt committed 30 days ago

Verified 5d43f737

[CI] fix url-encoding behavior in nightly metadata generation (#29787)

Harry-Chen committed 30 days ago

Verified 37593deb

[Doc] fix heading levels (#29783)

KKKZOZ committed 30 days ago

Verified f5516039

[CI] Renovation of nightly wheel build & generation (#29690)

Harry-Chen committed 31 days ago

Verified 36db0a35

[Bugfix] TypeError: 'NoneType' object is not callable (#29414)

mostrowskix committed 31 days ago

Verified 5cfa967e

[v1] Add real sliding window calculation to FlexAttention direct BlockMask building (#26015)

Isotr0py committed 31 days ago

Verified b95db244

[multimodal][test] Reduce memory utilization for test_siglip to avoid OOM (#29504)

zhxchen17 committed 31 days ago

Verified ad9d656b

[XPU] Fix AWQ skipped layer detection in IPEX quantization (#29774)

faaany committed 31 days ago

Verified f37e8938

[Misc] Unify tokenizer registration (#29767)

DarkLight1337 committed 31 days ago

Verified f0a28bf6

[crashfix] Eagle + multimodal can crash on mm cache miss (#29750)

mickaelseznec committed 31 days ago

Verified 86e178f7

[Frontend] Add tool filtering support to ToolServer (#29224)

daniel-salib committed 31 days ago

Verified 014ece97

[Frontend] Resettle pooling entrypoints (#29634)

noooop committed 31 days ago

Verified 62de4f42

[CI] Skip paddleocr_vl for transformer 4.57.3 (#29758)

hl475 committed 31 days ago

Verified 83805a60

Make PyTorch profiler gzip and CUDA time dump configurable (#29568)

zhangruoxu committed 31 days ago

Verified 1ab8fc81

[MoE] CuteDSL MoE with Nvfp4 DeepEP dispatch (#27141)

wenscarl committed 31 days ago

Verified f72a817b

[Model Runner V2] Use packed mask for prompt bin counts (#29756)

WoosukKwon committed 31 days ago

Verified ec38a736

[Misc]Remove redundant hidden_size property in ModelConfig (#29749)

charlotte12l committed 31 days ago

Verified 21c26279

[Feat] Support non-gated activations in NVFP4 modelopt path (#29004)

omera-nv committed 31 days ago

Verified 39d28108

Fix RoPE failures in Transformers nightly (#29700)

hmellor committed 31 days ago

Verified cd719de5

[ROCm][Attention] Sliding window support for `AiterFlashAttentionBackend` (#29234)

ganyi1996ppo committed 32 days ago

Verified 8c363ed6

[Core] Enable `inputs_embeds_size` separate from `hidden_size` (#29741)

DarkLight1337 committed 32 days ago

Verified 64bc09ba

[Bugfix] Fix mismatched nvfp4 gemm output shape (#29742)

Isotr0py committed 32 days ago

Verified 47539cfd

[Misc] Update `TokenizerLike` interface and move `get_cached_tokenizer` (#29730)

DarkLight1337 committed 32 days ago

Verified 2afcec4d

[Doc]: Fix typo in fused_moe layer (#29731)

BowTen committed 32 days ago

Verified 9381b5cd

[Bugfix][sleepmode][fp8 kv cache]: Fix FP8 KV cache + sleep(level=2) gibberish output (#28783)

Flink-ddd committed 32 days ago

Verified 66b58402

Fix AttributeError about _use_fi_prefill (#29734)

hl475 committed 32 days ago

Verified 82c795d6

[Quantization] Enable compressed-tensors AWQ for Turing GPU (#29732)

Isotr0py committed 32 days ago

Verified e1464c3a

[LoRA] Support FusedMoE LoRA Triton kernel for mxfp4 (#29708)

xyang16 committed 32 days ago

Verified a491b091

[Bugfix] Revert test_tokenization.py (#29729)

jeejeelee committed 32 days ago

Verified b9d0504a

[Kernel][Quantization] add w4a8 support for marlin kernel (#24722)

jinzhen-lin committed 32 days ago

Verified 1656ad37

Newer Older