Lucas Wilkinson 0ae43dbf8c

[Attention] add DCP support for FLASH_ATTN_MLA backend (#24453) ...

Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
Co-authored-by: Matthew Bonanni <mbonanni@redhat.com>

2025-09-10 17:19:26 +08:00

..

adapter_commons

[Misc] Add SPDX-FileCopyrightText (#19100)

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

assets

[CI/Build] Serve images used by multimodal tests through local HTTP Server (#23907)

2025-09-03 16:13:11 +08:00

attention

[Feature] Disallow FlashMLA on Blackwell (#24521)

2025-09-09 14:59:34 -04:00

benchmarks

[Benchmark] Add option to skip oversampling in benchmark (#24457)

2025-09-09 22:00:17 +00:00

compilation

[Flashinfer] Support Flashinfer TRTLLM FP8-qkv BF16/FP16-out Attention Kernel (#23647)

2025-09-08 20:53:07 -07:00

config

[Bugfix] Guard _may_reorder_batch for encoder-only models on CPU (#24319) (#24348)

2025-09-10 14:24:42 +08:00

core

[Doc]: fix typos in Python comments (#24001)

2025-08-31 08:21:59 +00:00

device_allocator

[Bugfix] Add strong reference to CUDA pluggable allocator callbacks (#23477)

2025-08-24 12:56:17 +08:00

distributed

[P/D] MultiConnector supports shutdown (#24425)

2025-09-09 21:24:11 -07:00

engine

[Bugfix] Update Run:AI Model Streamer Loading Integration (#23845)

2025-09-09 21:37:17 -07:00

entrypoints

Consolidate rendering parameters into RenderConfig dataclass (#24543)

2025-09-10 08:44:47 +00:00

executor

[P/D] Add a shutdown method to the Connector API (#22699)

2025-09-07 23:07:00 -07:00

inputs

[Core] Simplify and unify mm uuid handling & auto-generated mm hash overrides processing. (#24271)

2025-09-09 21:36:09 -07:00

logging_utils

[Log] Use a relative path in debug-level logs to distinguish files with identical names (#23846)

2025-09-09 16:46:35 -04:00

lora

[Doc]: fix typos in Python comments (#24093)

2025-09-02 21:05:45 -07:00

model_executor

[Model] Limit CPU threads for image transformations in InternVL to reduce cpu contention. (#24519)

2025-09-10 16:45:44 +08:00

multimodal

[Core] Simplify and unify mm uuid handling & auto-generated mm hash overrides processing. (#24271)

2025-09-09 21:36:09 -07:00

platforms

[Hardware][Apple-CPU] Enable native bfloat16 on Apple Silicon (M2 and later) (#24129)

2025-09-10 03:50:21 +00:00

plugins

[docs][misc] IOProcessor plugins fixes (#24046)

2025-09-01 09:17:41 -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

[gpt-oss] tool parser supports for /chat/completions [1/n] (#22386)

2025-09-04 20:39:12 -07:00

third_party

fix some typos (#24071)

2025-09-02 20:44:50 -07:00

transformers_utils

[Bugfix] Update Run:AI Model Streamer Loading Integration (#23845)

2025-09-09 21:37:17 -07:00

triton_utils

[Kernels] Add Flash Linear Attention Kernels (#24518)

2025-09-10 00:04:41 +08:00

usage

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

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

utils

[ROCm][Feature] Enable Pipeline Parallelism with Ray Compiled Graph on ROCm (#24275)

2025-09-09 23:27:35 +00:00

v1

[Attention] add DCP support for FLASH_ATTN_MLA backend (#24453)

2025-09-10 17:19:26 +08:00

vllm_flash_attn

…

worker

[P/D] Add a shutdown method to the Connector API (#22699)

2025-09-07 23:07:00 -07:00

__init__.py

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

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

_custom_ops.py

[Kernel] Support decode context parallelism on Blackwell with CUTLASS MLA (#24385)

2025-09-08 09:27:12 +08:00

_ipex_ops.py

[XPU] support Triton Attention backend on Intel GPU (#24149)

2025-09-04 20:41:08 +08:00

beam_search.py

[Chore][V0 Deprecation] Move LogProb to a separate file (#24055)

2025-09-01 12:07:53 -07:00

collect_env.py

[Misc] collect flashinfer version in collect_env.py (#24378)

2025-09-07 03:30:41 +00:00

connections.py

[Misc] Add SPDX-FileCopyrightText (#19100)

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

env_override.py

[distributed][rl] remove nccl cumem env var override (#24141)

2025-09-03 06:45:25 +00:00

envs.py

[Core] Use sha256 bytes instead of BlockHash to reduce GC overhead (#23673)

2025-09-08 21:34:37 -07:00

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

[Log] Use a relative path in debug-level logs to distinguish files with identical names (#23846)

2025-09-09 16:46:35 -04:00

logits_process.py

[Misc] Add SPDX-FileCopyrightText (#19100)

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

logprobs.py

[Chore][V0 Deprecation] Move LogProb to a separate file (#24055)

2025-09-01 12:07:53 -07:00

outputs.py

[Chore][V0 Deprecation] Move LogProb to a separate file (#24055)

2025-09-01 12:07:53 -07:00

pooling_params.py

Fix wrong truncate_prompt_tokens type hint (#22761)

2025-08-30 20:39:38 +00:00

py.typed

…

sampling_params.py

[Sampler] Support returning all prompt logprobs (#23868)

2025-09-07 19:34:31 -07:00

scalar_type.py

[Doc]: fix typos in Python comments (#24042)

2025-09-01 19:07:45 -07:00

scripts.py

[Misc] Add SPDX-FileCopyrightText (#19100)

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

sequence.py

[Doc]: fix typos in Python comments (#24042)

2025-09-01 19:07:45 -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