Commits vllm-project/vllm

[Bugfix][ROCm] Preserve MoE weight padding for unquantized Triton path (#46381)

umarkovi-amd committed 3 minutes ago

Verified 81bcced4

[Platform] Replace `torch.cuda.mem_get_info` with `torch.accelerator.get_memory_info` (#44825)

jikunshang committed 11 minutes ago

Verified fb42e521

PD disagg with Mooncake Connector: GDN support (Qwen3.5) and MLA support (Deepseek-V4-Flash) (#46807)

andakai committed 22 minutes ago

Verified 0feca7ff

[Bugfix] Raise VLLMValidationError for non-integer logit_bias keys (#46612)

muhammadfawaz1 committed 32 minutes ago

Verified 97b5ce5c

[ROCm][CI][Multimodal] Use ROCm-aware FA availability check for Unlimited-OCR (#47004)

AndreasKaratzas committed 48 minutes ago

Verified 42365140

[Rust Frontend] Start current wave for a stale DP FirstRequest (#46833)

blasrodri committed 1 hour ago

Verified e45c8a9f

[Bugfix] Use larger workspace size for Flashinfer MLA LSE (#47074)

wzhao18 committed 1 hour ago

Verified b153dd3f

[Bugfix][Rust Frontend] Reject prompt_logprobs for streaming generate (#46839)

reidliu41 committed 1 hour ago

Verified 930f8dc0

[Rust Frontend] Avoid LoRA registry scans without active LoRA requests (#47040)

reidliu41 committed 1 hour ago

Verified a16dbd5b

Secondary tier implementation for PD disaggregation (#42285)

liranschour committed 1 hour ago

Verified bec232a9

[LoRA] Add language-backbone LoRA support for MiniCPM-V 4.6 (#46740)

linitra24 committed 2 hours ago

Verified b5c9e1ac

[XPU][UT]Fix xpu pass_config.fuse_norm_quant assert issue (#46804)

Yejing-Lai committed 2 hours ago

Verified ae2c4f3d

[Bugfix] Propagate default stop_token_ids to per-request SamplingParams (#35076)

sriganesh123 committed 2 hours ago

Verified fca432e6

fix(config): reject negative max_logprobs (except -1) and long_prefill_token_threshold (#44070)

hclsys committed 2 hours ago

Verified af1ee8c4

[Bugfix][MLA] Fix LSE log-base mismatch in DCP + FlashInfer MLA decode (#47079)

GirasoleY committed 4 hours ago

Verified 5b4cb695

[ROCm][CI] Make tests/v1/shutdown an importable package (#47085)

peizhang56 committed 4 hours ago

Verified 9fc0c080

[ROCm][CI] Move LM Eval Large Models (8 GPUs) to mi300 pool (#47094)

peizhang56 committed 4 hours ago

Verified f2b5fabb

[Rust Frontend] Add static HTTPS and mTLS support for HTTP and gRPC (#45890)

tahsintunan committed 5 hours ago

Verified b8cb75b1

[GDN] Improve kkt kernel of CuteDSL prefill backend (#46346)

gau-nernst committed 5 hours ago

Verified 43916891

[Bugfix][Reasoning] Fix thinking_token_budget not enforced on re-entry after forced end (#43757)

ashwing committed 5 hours ago

Verified cda05ee8

[KVTransfer] MultiConnector: merge kv_transfer_params dicts across connectors (#46777)

deng451e committed 6 hours ago

Verified 77654d08

[Bug] Fix sparse attention issue for GLM5.2 non-torch compile path (#47083)

yewentao256 committed 8 hours ago

Verified 75698e60

[ROCm][CI] Use spawn around the threaded OTLP test (#47003)

AndreasKaratzas committed 9 hours ago

Verified 8632c884

[CI][Bugfix] Add cohere_melody to ROCm test requirements (#47072)

peizhang56 committed 9 hours ago

Verified c3734e83

[ROCm][DeepEP] Stabilize high-throughput DBO for DP+EP (#46990)

AndreasKaratzas committed 9 hours ago

Verified 53f7553f

[ROCm][CI] Make memory sampling less racy in tests and sleep mode (#45490)

AndreasKaratzas committed 9 hours ago

Verified 4eb22799

[ROCm][CI] Soft Fail `Spec Decode Ngram + Suffix` and `Entrypoints Integration (LLM)` AMD Mirrors (#47067)

micah-wil committed 9 hours ago

Verified ebcf511e

Fix FA4 dynamic_causal for full attention layers (#46659)

MatthewBonanni committed 9 hours ago

Verified 8fc1b2d0

Fix transient dependency issues caused by `requirements/common.txt` (#47015)

hmellor committed 9 hours ago

Verified 5316638a

[Model Runner V2] support mamba hybrid models align prefix cache (#42406)

izhuhaoran committed 9 hours ago

Verified 61ab70ec