Commits vllm-project/vllm

[Misc] Bump opencv-python dependecy version to 4.13 (#32668)

Isotr0py committed 86 days ago

Verified 444e2e7e

[Cleanup] Move scheduler `get_routed_experts` logic to separate method (#32706)

njhill committed 86 days ago

Verified bc14663e

[torch.compile] Improve Cold Start for MoEs (#32805)

zou3519 committed 86 days ago

Verified 654a71fc

[Misc][BE] Turn on strict type coverage for vllm/compilation (#31756)

Lucaskabela committed 87 days ago

Verified 15e302df

[Frontend] Introduce Renderer for processing chat messages (using `ModelConfig`) (#30200)

DarkLight1337 committed 87 days ago

Verified d117a4d1

OffloadingConnector: Support kernel_block_size != block_size (#30692)

orozery committed 87 days ago

Verified 421012b6

[Frontend] add prompt_cache_key for openresponses (#32824)

chaunceyjiang committed 87 days ago

Verified 841d53aa

[CI] refactor release pipeline config into groups (#32833)

Harry-Chen committed 87 days ago

Verified 1752262e

[Bugfix] Fix Whisper/encoder-decoder GPU memory leak (#32789)

NickLucche committed 87 days ago

Verified ea6102b8

[Frontend][2/n] Make pooling entrypoints request schema consensus | ChatRequest (#32574)

noooop committed 87 days ago

Verified 328cbb27

Enable Cross layers KV cache layout at NIXL Connector (#30207)

liranschour committed 87 days ago

Verified 64e3d67a

[Benchmark] Don't default to `temperature==0` in `vllm bench serve` (#32723)

njhill committed 87 days ago

Verified 098b2d66

[Misc] Replace urllib's `urlparse` with urllib3's `parse_url` (#32746)

Isotr0py committed 87 days ago

Verified 8ebf271b

[AMD][ROCm] MoRI EP: a high-performance all2all backend (#28664)

alexsun07 committed 87 days ago

Verified 49a12622

[Model] Extend `collect_children` and `no_init_weights` contexts (#32757)

DarkLight1337 committed 87 days ago

Verified 2b8a38b6

[bench] add start_times field to vllm bench serve json result (#32667)

kebe7jun committed 87 days ago

Verified 1bf1a34b

[ROCm][CI][Docs] Add comment explaining TRITON_ATTN fallback for ROCm (#32835)

AndreasKaratzas committed 87 days ago

Verified a8102998

[ROCm][CI] Fix AITER test flakiness by using explicit attention backend (#32346)

AndreasKaratzas committed 87 days ago

Verified eb1629da

[ROCm][CI] Lower Acceptance Len Threshold For test_draft_model_quantization (#32731)

micah-wil committed 87 days ago

Verified 019e2c3b

Upgrade transformers-4.57.5 (#32287)

huydhn committed 87 days ago

Verified f5fdec8c

[Llama.py -> mistral.py] Extract mistral-only relevant code into separate file (#32780)

patrickvonplaten committed 87 days ago

Verified 1579c9b5

[FlashMLA] Update FlashMLA to expose new arguments (#32810)

LucasWilkinson committed 87 days ago

Verified 889722f3

[ROCm][CI] fix get_valid_backends (#32787)

divakar-amd committed 87 days ago

Verified 49d96538

[Docs] Remove outdated async_scheduling limitation with speculative decoding (#32775)

ikaadil committed 87 days ago

Verified a1d82466

Cleanup some huggingface_hub-related stuff (#32788)

Wauplin committed 87 days ago

Verified 24a163ed

[EC Connector] Optimize remote cache check in scheduler (#32585)

knlnguyen1802 committed 87 days ago

Verified 378385b9

[Bugfix] Fix potential EAGLE spec decode segfault during graph capture (#32818)

mawong-amd committed 87 days ago

Verified c5487e2b

[Deprecation] Remove deprecated environment variables (#32812)

yewentao256 committed 87 days ago

Verified 6437ff1f

[Model Runner V2] Do not error on attention backends (#32820)

WoosukKwon committed 87 days ago

Verified 5e00b561

[Model Runner V2] Refactor Prompt Logprobs (#32811)

WoosukKwon committed 87 days ago

Verified 408195ec