vllm-project/vllm

Pull Requests Commits

[XPU] Enable external_launcher to serve as an executor via torchrun (#21021)

chaojun-zhang committed 142 days ago

Verified 25d585ab

[v1][sampler] Inplace logprobs comparison to get the token rank (#21283)

houseroad committed 143 days ago

Verified 8d0a01a5

[perf] Speed up align sum kernels (#21079)

hj-mistral committed 143 days ago

Verified 0ec82edd

Fix bad lm-eval fork (#21318)

mgoin committed 143 days ago

Verified 005ae9be

[DP] Fix Prometheus Logging (#21257)

robertgshaw2-redhat committed 143 days ago

Verified 29d1ffc5

[Attention] Clean up iRoPE in V1 (#21188)

LucasWilkinson committed 143 days ago

Verified 304dce7e

[Misc] Add dummy maverick test (#21199)

minosfuture committed 143 days ago

Verified 6ece16c4

[BugFix] make utils.current_stream thread-safety (#21252) (#21253)

simpx committed 143 days ago

Verified a0e827e0

[CPU] Enable shared-memory based pipeline parallel for CPU backend (#21289)

bigPYJ1151 committed 143 days ago

Verified a15a50fc

[Misc] Add sliding window to flashinfer test (#21282)

WoosukKwon committed 143 days ago

Verified 6dda13c8

Add Nvidia ModelOpt config adaptation (#19815)

Edwardf0t1 committed 143 days ago

Verified 6b46c4b6

[Misc] unify variable for LLM instance (#20996)

andyxning committed 143 days ago

Verified d9784107

[Docs] Make tables more space efficient in `supported_models.md` (#21291)

hmellor committed 143 days ago

Verified e6b90a28

[Docs] Fix hardcoded links in docs (#21287)

hmellor committed 143 days ago

Verified be54a951

[Model][1/N] Support multiple poolers at model level (#21227)

DarkLight1337 committed 143 days ago

Verified 042af0c8

[Bugfix] Fix missing placeholder in logger debug (#21280)

DarkLight1337 committed 143 days ago

Verified 378d33c3

Add the instruction to run e2e validation manually before release (#21023)

huydhn committed 143 days ago

Verified 940af1f0

[Docs] Add RFC Meeting to Issue Template (#21279)

simon-mo committed 143 days ago

Verified 92615d7f

[CI] Cleanup modelscope version constraint in Dockerfile (#21243)

yankay committed 143 days ago

Verified 8188196a

[bugfix] fix syntax warning caused by backslash (#21251)

1195343015 committed 144 days ago

Verified 7ba34b12

[Model] Support VLMs with transformers backend (#20543)

zucchini-nlp committed 144 days ago

Verified 9499e26e

[Model] use AutoWeightsLoader for bart (#18299)

calvin0327 committed 144 days ago

Verified 51ba8395

Enable v1 metrics tests (#20953)

eicherseiji committed 144 days ago

Verified d1fb65bd

[TPU] support fp8 kv cache quantization (#19292)

yaochengji committed 144 days ago

Verified 3a1d8940

[Docs] [V1] Update docs to remove enforce_eager limitation for hybrid models. (#21233)

tdoublep committed 145 days ago

Verified 2b504eb7

GLM-4 Update (#20736)

zRzRzRzRzRzRzR committed 145 days ago

Verified 10eb24cc

[BugFix] Fix full cuda graph slot_mapping (#21228)

fhl2000 committed 145 days ago

Verified 2e8cbb58

[V0 Deprecation] Deprecate BlockSparse Attention & Phi3-Small (#21217)

WoosukKwon committed 145 days ago

Verified 752c6ade

[V1] [Hybrid] Enable piecewise CUDA Graph for mamba layers (#21194)

tdoublep committed 145 days ago

Verified 881e3cbe

[BugFix] Make PD work with Ray (#21072)

kouroshHakha committed 145 days ago

Verified 9f414a12

Newer Older