Wentao Ye 394591e343

[Feature] Enable DeepGEMM Linear on B200; 1.5% E2E throughput improvement (#23351) ...

Signed-off-by: yewentao256 <zhyanwentao@126.com>

2025-08-21 21:01:08 -07:00

..

adapter_commons

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

assets

[Bugfix] Fix ndarray video color from VideoAsset (#21064)

2025-07-19 02:17:16 -07:00

attention

[Kernel] Add FP8 support with FlashMLA backend (#22668)

2025-08-22 02:26:32 +00:00

benchmarks

[BugFix] Fix Python 3.9 Support (#23306)

2025-08-20 23:23:56 -07:00

compilation

[torch.compile] Support conditional torch.compile per module (#22269)

2025-08-20 16:52:59 +00:00

config

[Core] Support custom executor qualname (#23314)

2025-08-22 09:40:54 +08:00

core

[V0 Deprecation] Remove multi-step scheduling (#22138)

2025-08-12 20:18:39 -07:00

device_allocator

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

distributed

[Feature] use --eplb_config to set eplb param (#20562)

2025-08-20 14:07:28 -07:00

engine

[Kernel] Add FP8 support with FlashMLA backend (#22668)

2025-08-22 02:26:32 +00:00

entrypoints

[Deprecation] Remove prompt_token_ids arg fallback in LLM.generate and LLM.embed (#18800)

2025-08-22 10:56:57 +08:00

executor

[Refactor] Define MultiModalKwargsItems separate from MultiModalKwargs (#23053)

2025-08-18 09:52:00 +00:00

inputs

[Multimodal] Always enable hashing mm data (#23308)

2025-08-21 07:23:28 -07:00

logging_utils

[BugFix] Don't catch BaseException when dumping execute_model errors (#19626)

2025-06-16 11:01:08 +00:00

lora

[Kernel/Quant] Remove the original marlin format and qqq (#23204)

2025-08-20 15:13:36 -04:00

model_executor

[Feature] Enable DeepGEMM Linear on B200; 1.5% E2E throughput improvement (#23351)

2025-08-21 21:01:08 -07:00

multimodal

[Multimodal] Always enable hashing mm data (#23308)

2025-08-21 07:23:28 -07:00

platforms

[Kernel] Add FP8 support with FlashMLA backend (#22668)

2025-08-22 02:26:32 +00:00

plugins

[XPU] upgrade torch 2.8 on for XPU (#22300)

2025-08-08 17:03:45 -07:00

profiler

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

ray

[Misc] Getting and passing ray runtime_env to workers (#22040)

2025-08-01 23:54:40 -07:00

reasoning

[Frontend] Avoid list copies in serving_chat.py (#22947)

2025-08-16 02:06:30 +00:00

third_party

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

transformers_utils

[Bugfix] Fix extra whitespace in strings caused by newline (#23272)

2025-08-20 22:03:00 -07:00

triton_utils

[Bugfix] Add missing field to TritonLanguagePlaceholder (#20812)

2025-07-11 05:25:11 -07:00

usage

[Misc] Add logger.exception for TPU information collection failures (#20510)

2025-07-05 07:24:32 -07:00

utils

[Feature] Enable DeepGEMM Linear on B200; 1.5% E2E throughput improvement (#23351)

2025-08-21 21:01:08 -07:00

v1

[Kernel] Add FP8 support with FlashMLA backend (#22668)

2025-08-22 02:26:32 +00:00

vllm_flash_attn

[BugFix] Fix vllm_flash_attn install issues (#17267)

2025-04-27 17:27:56 -07:00

worker

[Misc] Misc code cleanup/simplification (#23304)

2025-08-21 17:22:55 +00:00

__init__.py

[Perf][CLI] Improve overall startup time (#19941)

2025-06-22 23:11:22 +00:00

_custom_ops.py

[Kernel] Add FP8 support with FlashMLA backend (#22668)

2025-08-22 02:26:32 +00:00

_ipex_ops.py

[XPU]Fix flash_attn_varlen_func interface on xpu (#22350)

2025-08-06 19:28:11 -07:00

beam_search.py

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

collect_env.py

[Misc] Refactor: Improve argument handling for conda command (#20481)

2025-07-15 19:43:19 -07:00

connections.py

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

env_override.py

[Misc] Reduce warning message introduced in env_override (#19476)

2025-06-11 17:29:54 +08:00

envs.py

[Core] Always use tensor cores for Flashinfer Decode Wrapper (#23214)

2025-08-21 16:02:11 -04:00

forward_context.py

[Core] Allow full cudagraph with separate attention routines and orthogonal to compilation, add support for FA2 and FlashInfer (#20059)

2025-08-15 10:01:39 -04:00

logger.py

Optimize logger init performance by using module-level constants (#22373)

2025-08-06 20:32:19 -07:00

logits_process.py

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

outputs.py

[Model][2/N] Automatic conversion of CrossEncoding model (#19978)

2025-07-03 13:59:23 +00:00

pooling_params.py

Support token_type_ids in V1 with less code changes (#21985)

2025-08-10 22:54:59 -07:00

py.typed

…

sampling_params.py

[Docs] Fix warnings in docs build (#22588)

2025-08-10 05:49:51 -07:00

scalar_type.py

[Kernel] [Quantization] Add MXFP4 and bias support for marlin kernel (#22428)

2025-08-14 11:23:22 -07:00

scripts.py

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

sequence.py

[Bugfix] fix IntermediateTensors equal method (#23027)

2025-08-18 02:58:11 -07:00

tasks.py

[V1] Get supported tasks from model runner instead of model config (#21585)

2025-07-25 05:36:45 -07:00

test_utils.py

[Misc] Modify the organization of GLM series (#22171)

2025-08-03 23:51:20 -07:00

tracing.py

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

version.py

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00