Morph MORPH
SIGN IN SIGN UP
tejas / vllm

A high-throughput and memory-efficient inference and serving engine for LLMs

0 0 122 Python

COMMITS

main
October 23, 2025
J
[Chore] Remove duplicate `has_` functions in vllm.utils (#27372)
Jonathan Chen committed
e255d92
W
[Model] Add num_cached_tokens for PoolingRequestOutput (#27378)
wang.yuqi committed
3729ed0
G
[V1][spec decode] return logprobs for spec decoding (#26060)
Giancarlo Delfin committed
6644796
A
[CORE] Support Prefix Caching with Prompt Embeds (#27219)
Andrew Sansom committed
ff93cc8
P
[Bugfix][Core] running queue index leakage exception (#26754)
PiteXChen committed
243ed7d
F
[Bugfix] Fix incorrect kv cache metrics in grafana.json (#27133)
fangpings committed
7e09410
C
[Bugfix] Fix SLA tuner initialization (#27355)
Cyrus Leung committed
6738e4a
I
[Bugfix] Fix deepseek-ocr multi-image inference and add `merge_by_field_config=True` with tensor schema support (#27361)
Isotr0py committed
2566dca
October 22, 2025
M
[MLA] Bump FlashMLA (#27354)
Matthew Bonanni committed
b4fda58
D
[Chore] Separate out system utilities from vllm.utils (#27201)
dongbo910220 committed
a0003b5
D
[BugFix] bugfix for Flash Attention MLA with full cuda graph IMA following pr-25490 (#27128)
Daisy-Ma-coder committed
5beacce
R
[Feature] publisher default set zmq in kv_event config (#26915)
rongfu.leng committed
8669c69
S
[Prefix Cache] Use LoRA name for consistent KV-cache block hashing (#27211)
Sage committed
1651003
W
[Doc] Fix numbering sequence in prefix caching (#27357)
William Song committed
1cb8c6c
L
[Model] Revert PR #26715: Restore custom PaliGemma and Gemma3-MM impl… (#27309)
Luciano Martins committed
e05a675
I
[Bugfix] Disable FlexAttention direct block mask building for encoder-only models (#27344)
Isotr0py committed
084a9da
R
Support Anthropic API /v1/messages Endpoint (#22627)
RED committed
c9461e0
N
[P/D] Dynamic `kv_output_aggregator` collect size (#26734)
Nicolò Lucchesi committed
4dfdb82
R
[Frontend] Require flag for loading text and image embeds (#27204)
Russell Bryant committed
58fab50
I
[Bugfix] Fix HF format InternVL large variants video processing (#27330)
Isotr0py committed
db6f28d
C
[Bugfix] Make `get_mrope_input_positions` instance methods (#27342)
Cyrus Leung committed
14e2f12
C
[NIXL] use Host buffer to support TP_ratio > 1 for XPU (#27140)
Chendi.Xue committed
7c4767f
J
[Bugfix] Add missing 'is_internal_router' attribute to FusedMoEWithLoRA (#27351)
Jee Jee Li committed
9771e0b
R
[bugfix] remove unused parameters to reduce unnecessary vram usage (#26789)
Reinforce-II committed
980de31
W
[Bug] Fix DeepSeek-V2.5-1210-FP8 issue (#27267)
Wentao Ye committed
1c16084
M
[NIXL] Terminate handshake listener thread in shutdown (#26404)
Mark McLoughlin committed
4ca13a8
I
[Model] Upstream Deepseek-OCR model (#27247)
Isotr0py committed
675aa2e
D
[Chore] Separate out optional dependency checks from vllm.utils (#27207)
dongbo910220 committed
3ae082c
A
Mirroring changes in test-pipeline.yaml into test-amd.yaml (#27242)
Alexei-V-Ivanov-AMD committed
49c00fe
M
[docs] Update v1 metrics design doc (#27332)
Mark McLoughlin committed
141d3b9