vLLM/vllm at d125b540b71a6d406ff6ceed53aaa028005c7fe9 - vLLM

[Bugfix] Align OpenCV video metadata timeline (#47099 )

2026-06-30 20:43:42 +00:00

[Pooling] Report input throughput for batched requests (#53213 )

2026-08-22 09:01:40 -04:00

speculative decoding under tensor parallelism (TP>1) , workspace creation select max hidden dim of target and draft model (#52193 )

2026-08-24 21:33:36 +00:00

config

[Core] Disable fuse_allreduce_rms under VLLM_BATCH_INVARIANT (non-deterministic under TP) (#51292 )

2026-08-25 15:16:09 -04:00

cute_utils

[Kimi-K3] Add GEMM-RS for sequence parallelism (#52079 )

2026-08-13 09:02:28 -07:00

device_allocator

[Core] Add CuMemAllocator.discard() for tag-selective GPU memory release (#52514 )

2026-08-16 23:15:14 +08:00

distributed

[Bugfix][NIXL] Fix Mamba prefill truncation ordering (#53523 )

2026-08-24 19:13:27 -07:00

engine

[Model Runner V2] batch-sharded sample (#50465 )

2026-08-24 15:49:57 -07:00

entrypoints

[Bugfix][Frontend] Apply the stop string limit to Cohere requests (#53750 )

2026-08-25 16:41:08 +00:00

inputs

[Refactor] Remove InputPreprocessor (#53064 )

2026-08-20 08:39:48 +00:00

ir

[CI] Use explicit devices in IR tests (#49513 )

2026-07-23 20:08:14 -07:00

kernels

[Bugfix][Triton] Make fp8_min/fp8_max constexpr in _quantize_pad_fp8_kernel (#51872 )

2026-08-12 03:09:54 +00:00

logging_utils

…

lora

[Model Runner V2] batch-sharded sample (#50465 )

2026-08-24 15:49:57 -07:00

model_executor

[Perf] Autotune batch invariance triton kernel in blackwell, 33.6% E2E latency reduction (#53649 )

2026-08-25 15:06:01 -04:00

models

[Kimi K3][Kernel] Enable low-latency decode GEMM dispatch on SM100 (#53534 )

2026-08-25 16:32:28 -07:00

multimodal

[Bugfix][Multimodal] Reject malformed base64 audio with 400 instead of 500 (#53744 )

2026-08-25 17:31:42 +00:00

parser

[Bugfix] Handle parenthesized Gemma4 tool calls (#53657 )

2026-08-25 08:21:42 -04:00

platforms

[Bugfix] Record non-ImportError attention backend probe failures instead of crashing engine init (#51703 )

2026-08-20 02:36:00 -07:00

plugins

[MISC] Cleanup deprecated parameters (#53559 )

2026-08-24 10:19:04 -07:00

profiler

[Profiler] Fix start_profile permanently no-op after max_iterations auto-stop (#51839 )

2026-08-25 08:13:29 -04:00

ray

Propagate Flash Attention cache configuration to Ray workers (#49177 )

2026-07-21 07:47:53 +00:00

reasoning

Fix Cohere ChatV2 citation and tool handling issues (#52175 )

2026-08-21 15:23:17 -07:00

renderers

[MM] Simplify prompt updates: replace PromptSeq with list[int] (#53372 )

2026-08-23 14:50:32 +08:00

third_party

[Core] Check for GPU<->CPU syncs during CI (#43107 )

2026-08-14 17:42:05 -07:00

tilelang_utils

[ROCm] Defer tilelang import through its import from vllm.tilelang_utils import tilelang and relaxed has_tilelang (#51159 )

2026-08-13 13:12:40 -05:00

tokenizers

[Bugfix][Tokenizer] Replace bare asserts in the DeepSeek V4 encoder (#53747 )

2026-08-25 16:30:23 +00:00

tool_parsers

[Frontend] Prevent Kimi K3 reserved markers in response text (#52889 )

2026-08-21 09:06:52 -07:00

tracing

…

transformers_utils

[Model] Migrate FlexOlmo, Olmo3 and Hunyuan V1/VL to the Transformers modeling backend (#53615 )

2026-08-24 22:39:05 -07:00

triton_utils

Exclude the cpu backend from vLLM's active-Triton-driver count (#53530 ) (#53530 )

2026-08-25 04:13:13 +00:00

usage

[Bugfix] Close usage telemetry HTTP sessions (#51219 )

2026-08-08 00:29:36 +00:00

utils

[Bugfix] Resolve B12X modules before Dynamo tracing (#53326 )

2026-08-25 09:20:15 -07:00

v1

[Kimi Bug] Fix k3 torch.AcceleratorError: CUDA error: an illegal memory access was encountered (#53773 )

2026-08-25 17:38:48 -07:00

vllm_flash_attn

Revert "[Kernel] Gemma-4 FA4 FP8 Kernel" (#52987 )

2026-08-19 10:48:46 -07:00

__init__.py

…

_aiter_ops.py

[ROCm][Perf] Fuse DeepSeek-V4 mHC post/pre and RMSNorm with AITER (#52737 )

2026-08-20 04:51:04 +00:00

_custom_ops.py

[CPU] Add AMX-only high-performance MLA backend for DeepSeek V2/V3/R1 (#52616 )

2026-08-19 10:28:52 +08:00

_xpu_ops.py

[Bugfix][XPU] Run GDN attention as eager break under breakable cudagraph (#51928 )

2026-08-12 16:23:39 +08:00

collect_env.py

[XPU] fix collecting oneccl version info (#47104 )

2026-08-04 11:53:28 +00:00

connections.py

Reject oversized media before fully downloading it (#51896 )

2026-08-24 01:15:09 +00:00

env_override.py

[Perf] Add VLLM_TRITON_FORCE_FIRST_CONFIG to skip Triton autotuning (#42425 )

2026-06-16 15:16:45 +02:00

envs.py

fix(security): enforce VLLM_MAX_AUDIO_CLIP_FILESIZE_MB on all audio paths (#53561 )

2026-08-24 21:37:02 +00:00

exceptions.py

[Frontend][Core] Standardize request error handling with VLLMError hierarchy (#49665 )

2026-07-29 04:20:40 +00:00

forward_context.py

[Feature] Support sequence parallel without the need for DP, 1.9%~5.0% E2E Throughput Improvement (#47070 )

2026-07-05 05:41:30 -07:00

logger.py

…

logits_process.py

…

logprobs.py

…

model_inspection.py

…

outputs.py

[Frontend][Core][Spec Decode] Per-request acceptance stats in OpenAI API responses (#48915 )

2026-08-20 15:42:49 -07:00

pooling_params.py

[MISC] Cleanup deprecated parameters (#53559 )

2026-08-24 10:19:04 -07:00

py.typed

…

sampling_params.py

[Feature][DSpark]: Logprobs adaptive verification (#52242 )

2026-08-25 10:22:33 -04:00

scalar_type.py

Humming support for 2/3/5/6/7-bit pack-quantized weight-only inference (#46389 )

2026-06-24 13:53:54 -04:00

scripts.py

…

sequence.py

…

tasks.py

[Frontend] Move api_server.py out openai folder (#52131 )

2026-08-19 09:43:46 +00:00

version.py

…