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 24, 2025
A
[Chore] remove structural tags logging lines (#27451)
Aaron Pham committed
d4c574c
U
Fix EventPublisherFactory logic for disabled KV cache events (#27419)
usberkeley committed
c528b90
F
[Bugfix][CI] Move resolving cudagraph_mode before initializing attn_metadata_builder (#27427)
fhl2000 committed
85fee74
H
[Misc] Add TPU usage report when using tpu_inference. (#27423)
hfan committed
8dbe0c5
X
[Kernel] Add GPTQv2 format support for low-bit or asymmetric quantization, by adapting gptq_gemm (#26092)
Xiangyu Li committed
5cc6bdd
H
Fix pooling adapters for Transformers backend (#27338)
Harry Mellor committed
1f9460c
X
Granite 4.0 quark quantization support (#26944)
xiao-llm committed
70022ff
October 23, 2025
A
[Hardware][POWERPC] Disable oneDNN path in vllm/model_executor/layers/utils.py for Powerpc (#27422)
Akash kaothalkar committed
f417746
Y
[Model] Siglip Embedding Support (#27324)
Yu Jiaqi committed
0552cfb
K
[Bugfix][DP] Fix creating too many DP Placement Groups (#26880)
Kebe committed
51dd14a
M
[Attention] Fix FlashMLA metadata builder arguments for q_len > 1 (#27368)
Matthew Bonanni committed
dbfbf9f
J
[Chore] Separate out `vllm.utils.platform_utils.py` (#27374)
Jonathan Chen committed
ca76486
V
[Misc] Add triton_kernels dependency (#27370)
Varun Sundar Rabindranath committed
a9f55dc
I
[Bugfix] Fix AWQ marlin layer skipping (#27416)
Isotr0py committed
81d5bb7
G
[Bugfix][ROCm][DeepSeek] Fix for forward_hip in rope for DeepSeek (#27373)
Gregory Shtrasberg committed
0825197
A
[Bugfix] Fix dp_chunking enablement logic in FusedMoE layer (#27220)
Alexander Matveev committed
9ef3d5b
A
Mirroring the test definitions (2025-10-22) (#27362)
Alexei-V-Ivanov-AMD committed
295c7f0
W
[Frontend][4/N] Improve all pooling task | Add plugin pooling task (#26973)
wang.yuqi committed
3fa2c12
C
[CI/Build] Remove unnecessary flags from test registry (#27353)
Cyrus Leung committed
fe2016d
I
[Misc] Remove use of CUDA_VISIBLE_DEVICES for device selection (fix DP slow startup time &c) (#26709)
Ilya Markov committed
237cf6d
N
[Feature] Pydantic validation for speculative.py (#27156)
Navya Srivastava committed
faee3cc
B
[Bugfix] Honor --mm_encoder_attn_backend when used (#27124)
Bradley D committed
570c3e1
H
Run mypy on the lowest supported Python version instead of system Python (#27048)
Harry Mellor committed
3a4255c
T
[Model] Add MoE support for NemotronH (#25863)
tomeras91 committed
6108946
T
[Metrics] [KVConnector] Add connector prefix cache hit rate stats (#26245)
Tova Movshovitz committed
88afa11
C
[CI] Reorganize entrypoints tests (#27403)
Chauncey committed
d00ce29
L
add SLA information into comparison graph for vLLM Benchmark Suite (#25525)
Louie Tsai committed
3b7bdf9
Z
[CI/Build] Fix AMD CI: test_cpu_gpu.py (#27388)
Zhewen Li committed
50b788a
L
[Bugfix] Fix args settings for guided decoding args (#27375)
Lucia Fang committed
fc059c7
C
[CI/Build] Fix Prithvi plugin test (#27393)
Cyrus Leung committed
bfb240c