vllm-project/vllm

Pull Requests Commits

Fix pytest verbosity for prime-rl ci

mgoin committed 84 days ago

Verified 38cf8237

[Misc][DP] support customized aggregated logger for dp (#24354)

luccafong committed 84 days ago

Verified 8317f723

Add tests for chunked prefill and prefix cache with causal pooling models (#26526)

maxdebayser committed 84 days ago

Verified d8bebb00

[ResponseAPI] Further polish message serialization and unit tests (#26728)

Jialin committed 84 days ago

Verified 35bc22f2

Pruning kernel Core Tests (#26727)

kfhfar committed 84 days ago

Verified fa96fb9c

[FrontEnd] UNREVERT CompilationConfig overhaul (#20283): deprecate use_inductor in favor of backend, simplify custom_ops (#26502)

morrison-turnansky committed 84 days ago

Verified e3fdb627

[Bug] Fix Assertion error DeepEP/csrc/kernels/intranode.cu:928: 'false and Unsupported type' (#26532)

yewentao256 committed 84 days ago

Verified 7200a21c

[CI Perf]Prune Tests in kernel/mamba (#26538)

kfhfar committed 84 days ago

Verified 577c72a2

[CI] Fix mypy for `vllm/distributed` (#26593)

yewentao256 committed 84 days ago

Verified 314285d4

[Frontend][1/N] Improve all pooling task | Support FP16 Embedding Base64 (Still uses fp32 by default). (#26414)

noooop committed 85 days ago

Verified d2a79385

[Quantization] [Performance] Enable Marlin GEMM kernels for the calibration-free RTN-based quantization (#26051)

sakogan committed 85 days ago

Verified 89342ce4

[CI][Release][Arm64]: Build arm64 release for gpu arch 8.9 (#26698)

cyb70289 committed 85 days ago

Verified f89f5993

[Log] Optimize Startup Log (#26601)

yewentao256 committed 85 days ago

Verified e251e457

[Model] Use merge_by_field_config for MM models (M-N) (#26710)

DarkLight1337 committed 85 days ago

Verified afc47e4d

[Bugfix][Speculative Decoding] Extend Eagle quantization config fix to llama_eagle.py (#26590)

rahul-tuli committed 85 days ago

Verified e3b90c1b

[Bugfix][Rocm] fix qr error when different inp shape (#25892)

haoyangli0109 committed 85 days ago

Verified 134f70b3

[CI/Build] upgrade compressed-tensors to 0.12.2 to address LGPLv3 (#26501)

csy1204 committed 85 days ago

Verified a1b2d658

[Misc] Separate prompt logging to debug (#26713)

aitsvet committed 85 days ago

Verified 5c7fe254

[P/D] [NixlConnector] kv load recovery integration (#26171)

wseaton committed 85 days ago

Verified 53c9a7ce

[UX] Speedup DeepGEMM warmup with heuristics (#25619)

mgoin committed 85 days ago

Verified 0d21b9b5

[FEATURE]: Use pydantic validation in `multimodal.py` config (#26629)

andycandy committed 85 days ago

Verified 10214b69

[Hardware][CPU] Disable torch.compile for RISC-V to prevent APIError (#26693)

ihb2032 committed 85 days ago

Verified 4a61950f

[unrevert] Add batch invariant kernel override for FlashInfer backend [2/n] (#26373)

bwasti committed 85 days ago

Verified 32637990

[Bugfix] Fix out of bound index issue for Jina-embedding-v3 RoPE with cuda graph (#26687)

Isotr0py committed 85 days ago

Verified 8e67b255

[ResponseAPI] Simplify input/output message serialization (#26620)

Jialin committed 85 days ago

Verified 4073c82c

[Model][0/N] Improve all pooling task | clean up (#25817)

noooop committed 85 days ago

Verified 767c3ab8

Ignore large reformatting PRs in `git blame` (#26690)

hmellor committed 85 days ago

Verified 4f207c71

[Model] Add reasoning_parser and tool_parser for Ernie45 thinking (#25027)

CSWYF3634076 committed 85 days ago

Verified 782505ed

[Model] Fix Skywork R1V mlp (#26673)

jeejeelee committed 85 days ago

Verified 98f30b8c

docs: wrong command in structured_outputs README (#26677)

yihong0618 committed 85 days ago

Verified 3cd36660

Older