quanliu 5dcd593baf

[Feature] Batch-Invariant Support for FA2 and LoRA (#30018) ...

Signed-off-by: quanliu <18646313696@163.com>
Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>

2025-12-09 10:01:38 -05:00

..

basic_correctness

[ROCm][CI] Fix test_max_len.py for Rocm (#29916)

2025-12-08 16:58:30 -05:00

benchmarks

Fix boolean nested params, add dict format support, and enhance plotting for vllm bench sweep (#29025)

2025-12-02 20:40:56 +00:00

compile

Mark qwen2_5_vl as xfail (#30283)

2025-12-09 01:14:10 +00:00

config

[torch.compile] caching of config fields should be opt-out by default (#26468)

2025-11-19 06:13:54 -08:00

cuda

…

detokenizer

…

distributed

[DCP][Bugfix][CI] Fix accuracy issue of DCP when using FLASH_ATTN_MLA (#30309)

2025-12-09 08:22:14 +00:00

engine

[Frontend] Remap -O to -cc commandline flag (#29557)

2025-11-28 21:51:12 +00:00

entrypoints

[Frontend] Binary embedding response does not return metadata by setting encoding_format to bytes_only. (#30249)

2025-12-08 12:01:21 +00:00

evals

…

kernels

[Kernel]Support W4A8 Grouped GEMM on Hopper (#29691)

2025-12-08 19:29:06 -08:00

lora

Lora MoE Align Improvements (#29257)

2025-12-09 10:35:16 +08:00

model_executor

Revert "[Renderer] Separate out RendererConfig from ModelConfig (#30145)" (#30199)

2025-12-07 00:00:22 -08:00

models

[Model][7/N] Improve all pooling task | Deprecation as_reward_model. Extract hidden states prefer using new multi-vector retrieval API (#26686)

2025-12-08 08:10:09 +00:00

multimodal

Revert "[Renderer] Separate out RendererConfig from ModelConfig (#30145)" (#30199)

2025-12-07 00:00:22 -08:00

plugins

[v1] Add PrefixLM support to FlexAttention backend (#27938)

2025-12-07 15:51:36 +00:00

plugins_tests

[Frontend] Resettle pooling entrypoints (#29634)

2025-12-01 15:30:43 +08:00

prompts

…

quantization

[Bugfix] Make compressed-tensors MoEs respect ignored layers (#28878)

2025-11-26 21:35:13 -05:00

reasoning

[Structured Output][Reasoning] Improves decoding throughput for models using single-token reasoning endings. (#30056)

2025-12-09 18:54:08 +08:00

rocm/aiter

…

samplers

[Core] Switch Flat logprob control from environment variable to SamplingParams (#28914)

2025-11-19 02:10:02 +00:00

standalone_tests

[CI] fix docker image build by specifying merge-base commit id when downloading pre-compiled wheels (#29930)

2025-12-03 14:08:19 -08:00

system_messages

…

tokenizers_

Add Mistral Large 3 and Ministral 3 (#29757)

2025-12-02 10:29:00 +00:00

tool_use

[BUGFIX] Mistral tool call parser v11+ (#30332)

2025-12-09 14:55:38 +00:00

tools

…

tpu

…

transformers_utils

[Chore]: Reorganize gguf utils funtions under transformers_utils (#29891)

2025-12-02 17:33:23 +00:00

utils_

[Frontend] Remove confusing -O.xx flag error (#30169)

2025-12-07 02:53:42 +00:00

v1

[Feature] Batch-Invariant Support for FA2 and LoRA (#30018)

2025-12-09 10:01:38 -05:00

vllm_test_utils

…

weight_loading

[ROCm][CI] Fix Weight Loading With Multiple GPU Tests on ROCm (#28984)

2025-11-19 21:31:33 +00:00

__init__.py

…

ci_envs.py

…

conftest.py

Support tokenization_kwargs override (#29794)

2025-12-06 09:12:53 +00:00

test_config.py

[Model][7/N] Improve all pooling task | Deprecation as_reward_model. Extract hidden states prefer using new multi-vector retrieval API (#26686)

2025-12-08 08:10:09 +00:00

test_embedded_commit.py

…

test_envs.py

[Bugfix] Respect VLLM_CONFIGURE_LOGGING value (#28671)

2025-12-03 22:00:52 +00:00

test_inputs.py

Revert "[Renderer] Separate out RendererConfig from ModelConfig (#30145)" (#30199)

2025-12-07 00:00:22 -08:00

test_logger.py

[BugFix] respect VLLM_LOGGING_LEVEL in logger (#29761)

2025-12-02 07:49:16 +00:00

test_logprobs.py

[Core] Switch Flat logprob control from environment variable to SamplingParams (#28914)

2025-11-19 02:10:02 +00:00

test_outputs.py

…

test_pooling_params.py

…

test_regression.py

…

test_routing_simulator.py

[MoE][Refactor] Make select_experts a non-static method (#29067)

2025-11-24 13:38:04 -05:00

test_scalartype.py

…

test_seed_behavior.py

…

test_sequence.py

…

test_triton_utils.py

…

test_version.py

…

test_vllm_port.py

…

utils.py

[ROCm][CI] Fix test_max_len.py for Rocm (#29916)

2025-12-08 16:58:30 -05:00