xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-09 23:15:31 +08:00
Code Issues Packages Projects Releases Wiki Activity
9,989 Commits 155 Branches 89 Tags
Commit Graph

121 Commits

Author SHA1 Message Date
Adrian Abeyta
c42ff4f4fd
[BugFix][torch.compile] KV scale calculation issues with FP8 quantization (#25513) 
Signed-off-by: adabeyta <aabeyta@redhat.com>
 2025-09-29 15:52:04 -04:00
fhl2000
f075693da7
[V1] address post issues related to #20059 (part 1) (#23046) 
Signed-off-by: fhl2000 <63384265+fhl2000@users.noreply.github.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
 2025-09-26 15:58:19 -04:00
Matthew Bonanni
3468f17ebe
[V0 deprecation] Remove _VLLM_V1 suffixes from attention backend names (#25489) 
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
Signed-off-by: Matthew Bonanni <mbonanni001@gmail.com>
 2025-09-25 17:37:50 +00:00
Luka Govedič
d5e0fca264
[torch.compile] Cleanup compilation tests and custom passes, add debug utils, fix DCE bug (#23091), fix test (#24376), and prep for custom op matching (#24604) (#24542) 
Signed-off-by: Luka Govedič <lgovedic@redhat.com>
Signed-off-by: luka <lgovedic@redhat.com>
Signed-off-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
 2025-09-22 12:30:05 -07:00
Daisy-Ma-coder
cfbee3d0e7
[CLI env var] Add VLLM_FLASH_ATTN_MAX_NUM_SPLITS_FOR_CUDA_GRAPH in env variables (#25274) 
Signed-off-by: qqma <qqma@amazon.com>
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Co-authored-by: qqma <qqma@amazon.com>
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>
 2025-09-22 10:37:43 -07:00
Woosuk Kwon
bc6e542d9f
Remove V0 attention backends (#25351) 
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
 2025-09-21 16:03:28 -07:00
Woosuk Kwon
52c2a8d4ad
[V0 Deprecation] Remove LLMEngine (#25033) 
Signed-off-by: Woosuk Kwon <woosuk@thinkingmachines.ai>
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
 2025-09-20 17:56:30 -07:00
Boyuan Feng
8945b001db
[torch.compile] CUDAGraph Inductor partition integration (#24281) 
Signed-off-by: Boyuan Feng <boyuan@meta.com>
Signed-off-by: Boyuan Feng <fby.1994@gmail.com>
Signed-off-by: boyuanfeng <boyuan@meta.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
 2025-09-20 01:02:15 +00:00
Lucas Kabela
3da17c2cc2
[Bugfix] Remove VLLM_TEST_DYNAMO_FULLGRAPH_CAPTURE #2969 (#25090) 
Signed-off-by: Lucas Kabela <lucaskabela@meta.com>
 2025-09-19 20:27:21 -04:00
elvischenv
e6585ddb45
[Bugfix] Fix accuracy issue for silu_mul + nvfp4 quant fusion kernel (#24833) 
Signed-off-by: elvischenv <219235043+elvischenv@users.noreply.github.com>
Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>
 2025-09-17 16:37:23 -07:00
Jiangyun Zhu
5bcc153d7b
[Compile] Fix noop_elimination pass and add tests for noop_elimination (#24880) 
Signed-off-by: zjy0516 <riverclouds.zhu@qq.com>
 2025-09-15 23:33:18 +00:00
co63oc
e26fef8397
fix some typos (#24616) 
Signed-off-by: co63oc <co63oc@users.noreply.github.com>
 2025-09-11 10:48:46 -07:00
Gregory Shtrasberg
9a161307f5
[torch.compile][ROCm][V1] Enable attention output FP8 fusion for V1 attention backends (#19767) 
Signed-off-by: Gregory Shtrasberg <Gregory.Shtrasberg@amd.com>
Signed-off-by: Luka Govedič <lgovedic@redhat.com>
Co-authored-by: Luka Govedič <lgovedic@redhat.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
 2025-09-10 13:59:55 -07:00
Jiangyun Zhu
b8a93076d3
[CI] execute all piecewise compilation tests together (#24502) 
Signed-off-by: zjy0516 <riverclouds.zhu@qq.com>
 2025-09-09 11:05:25 -07:00
Matthew Bonanni
620db1fc58
[Attention] FlashAttention MLA cudagraph support (#23958) 
Signed-off-by: Matthew Bonanni <mbonanni001@gmail.com>
Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>
 2025-09-08 22:05:26 +00:00
Jiangyun Zhu
7be141b2c5
[CI] Enable encoder model compilation test (#24442) 
Signed-off-by: zjy0516 <riverclouds.zhu@qq.com>
 2025-09-08 11:48:06 -07:00
elvischenv
e68dc2f014
[Bugfix] Fix unstable silu_mul+nvfp4 quant fusion test (#24370) 
Signed-off-by: elvischenv <219235043+elvischenv@users.noreply.github.com>
 2025-09-06 20:39:34 +00:00
elvischenv
eedb2a2a10
[Bugfix] Fix silu_mul+quant fusion test (#24341) 
Signed-off-by: elvischenv <219235043+elvischenv@users.noreply.github.com>
 2025-09-05 20:13:42 +00:00
nvjullin
37241077d5
[Misc] Removed force_fp8_e4m3fnuz from FP8LinearOp (#23725) 
Signed-off-by: Julien Lin <jullin@nvidia.com>
Signed-off-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
 2025-09-04 09:25:40 -04:00
co63oc
1bd007f234
fix some typos (#24071) 
Signed-off-by: co63oc <co63oc@users.noreply.github.com>
 2025-09-02 20:44:50 -07:00
elvischenv
16a45b3a28
[NVIDIA] Support SiluMul + NVFP4 quant fusion (#23671) 
Signed-off-by: jindih <jindih@nvidia.com>
Signed-off-by: elvischenv <219235043+elvischenv@users.noreply.github.com>
Co-authored-by: jindih <jindih@nvidia.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: Luka Govedic <lgovedic@redhat.com>
 2025-08-28 19:36:50 +00:00
nvjullin
f66673a39d
[Kernel] Added flashinfer fp8 per-tensor gemms (#22895) 
Signed-off-by: Julien Lin <jullin@nvidia.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
 2025-08-26 06:54:04 -07:00
elvischenv
24d0c9e6ed
[NVIDIA][torch.compile] Support Flashinfer TRTLLM FP8-q/kv NVFP4-out Attention Kernel (#22703) 
Signed-off-by: elvischenv <219235043+elvischenv@users.noreply.github.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
 2025-08-22 22:09:05 +00:00
Michael Goin
0cdbf5e61c
[Kernel/Quant] Remove the original marlin format and qqq (#23204) 
Signed-off-by: mgoin <mgoin64@gmail.com>
 2025-08-20 15:13:36 -04:00
Yong Hoon Shin
dfd2382039
[torch.compile] Support conditional torch.compile per module (#22269) 
Signed-off-by: Yong Hoon Shin <yhshin@meta.com>
 2025-08-20 16:52:59 +00:00
Michael Goin
d46d417b58
[CI Perf] Only test bfloat16 for tests/compile/test_fusion_all_reduce.py (#23132) 
Signed-off-by: mgoin <mgoin64@gmail.com>
 2025-08-19 20:18:52 -06:00
elvischenv
03752dba8f
[NVIDIA] Support Flashinfer TRTLLM FP8-q/kv/out Attention Kernel (#21716) 
Signed-off-by: elvischenv <219235043+elvischenv@users.noreply.github.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
 2025-08-19 08:22:15 -04:00
Woosuk Kwon
14006840ea
[V0 Deprecation] Remove V0 FlashInfer attention backend (#22776) 
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
 2025-08-18 19:54:16 -07:00
Michael Goin
4fc722eca4
[Kernel/Quant] Remove AQLM (#22943) 
Signed-off-by: mgoin <mgoin64@gmail.com>
Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>
 2025-08-16 19:38:21 +00:00
fhl2000
74f441f4b5
[Core] Allow full cudagraph with separate attention routines and orthogonal to compilation, add support for FA2 and FlashInfer (#20059) 
Signed-off-by: fhl <2410591650@qq.com>
Signed-off-by: fhl2000 <63384265+fhl2000@users.noreply.github.com>
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Signed-off-by: Lucas Wilkinson <LucasWilkinson@users.noreply.github.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Co-authored-by: Lucas Wilkinson <lwilkins@redhat.com>
Co-authored-by: Lucas Wilkinson <LucasWilkinson@users.noreply.github.com>
 2025-08-15 10:01:39 -04:00
Wentao Ye
5c3fbfe46b
[Feature] Full Cuda Graph Support for Cutlass MLA and 6% E2E Throughput Improvement (#22763) 
Signed-off-by: yewentao256 <zhyanwentao@126.com>
 2025-08-15 06:27:30 +00:00
Michael Goin
88faa466d7
[CI] Initial tests for SM100 Blackwell runner (#21877) 
Signed-off-by: mgoin <mgoin64@gmail.com>
 2025-08-01 16:18:38 -07:00
Ilya Markov
6e672daf62
Add FlashInfer allreduce RMSNorm Quant fusion (#21069) 
Signed-off-by: ilmarkov <imarkov@redhat.com>
Signed-off-by: ilmarkov <markovilya197@gmail.com>
Co-authored-by: ilmarkov <imarkov@redhat.com>
 2025-07-31 13:58:38 -07:00
cascade
287f527f54
[Feature] Add async tensor parallelism for scaled mm (#20155) 
Signed-off-by: cascade812 <cascade812@outlook.com>
 2025-07-30 17:23:41 -04:00
Richard Zou
04e38500ee
[Bugfix] VLLM_V1 supports passing other compilation levels (#19340) 
Signed-off-by: Richard Zou <zou3519@gmail.com>
 2025-07-29 09:35:58 -04:00
Cyrus Leung
86ae693f20
[Deprecation][2/N] Replace --task with --runner and --convert (#21470) 
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
 2025-07-27 19:42:40 -07:00
Yong Hoon Shin
4ac7713e32
Add test case for compiling multiple graphs (#21044) 
Signed-off-by: Yong Hoon Shin <yhshin@meta.com>
 2025-07-23 11:00:47 -07:00
Ilya Markov
37a7d5d74a
[Misc] Refactor AllReduceFusionPass. Remove parameter (#20918) 
Signed-off-by: ilmarkov <imarkov@redhat.com>
Co-authored-by: ilmarkov <imarkov@redhat.com>
 2025-07-15 06:57:40 +00:00
Richard Zou
ba8c300018
[BugFix] VLLM_DISABLE_COMPILE_CACHE=1 should disable all reads and writes from the cache (#20942) 
Signed-off-by: Richard Zou <zou3519@gmail.com>
 2025-07-15 01:26:18 +00:00
Ilya Markov
fc0f41d10a
Integration SM100 FlashInfer fused allreduce RMSNorm (#20691) 
Signed-off-by: ilmarkov <imarkov@redhat.com>
Co-authored-by: ilmarkov <imarkov@redhat.com>
 2025-07-11 18:58:15 -07:00
Luka Govedič
762be26a8e
[Bugfix] Upgrade depyf to 0.19 and streamline custom pass logging (#20777) 
Signed-off-by: Luka Govedic <lgovedic@redhat.com>
Signed-off-by: luka <lgovedic@redhat.com>
 2025-07-11 00:15:22 -07:00
Luka Govedič
31d5c1797f
[Perf][fp8] Use CustomOp abstraction for fp8 quant for better perf (#19830) 
Signed-off-by: Luka Govedic <lgovedic@redhat.com>
Co-authored-by: mgoin <mgoin64@gmail.com>
 2025-07-11 04:56:28 +00:00
Jee Jee Li
1caca5a589
[Misc] Add SPDX-FileCopyrightText (#20428) 
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
 2025-07-04 07:40:42 +00:00
Boyuan Feng
c01d1c5aba
use .dev for version comparison with pytorch nightly release (#20031) 
Signed-off-by: Boyuan Feng <boyuan@meta.com>
 2025-06-24 21:52:16 +00:00
cascade
e6327c9b3e
[Feature] Support sequence parallelism for static fp8 quantization (#19181) 
Signed-off-by: cascade812 <cascade812@outlook.com>
 2025-06-23 16:09:02 -04:00
Maximilien de Bayser
799397ee4f
Support embedding models in V1 (#16188) 
Signed-off-by: Max de Bayser <mbayser@br.ibm.com>
Signed-off-by: Max de Bayser <maxdebayser@gmail.com>
Signed-off-by: 22quinn <33176974+22quinn@users.noreply.github.com>
Co-authored-by: 22quinn <33176974+22quinn@users.noreply.github.com>
 2025-06-18 21:36:33 -07:00
Richard Zou
ed33349738
[BugFix] Fix use_cudagraph=False (#19612) 
Signed-off-by: Richard Zou <zou3519@gmail.com>
 2025-06-19 08:23:12 +08:00
Charlie Fu
a44b1c951d
[Feature][ROCm] Add full graph capture support for TritonAttentionBackend (#19158) 
Signed-off-by: charlifu <charlifu@amd.com>
 2025-06-17 17:03:06 -04:00
Luka Govedič
3597b06a4f
[CUDA] Enable full cudagraph for FlashMLA (#18581) 
Signed-off-by: luka <luka@neuralmagic.com>
 2025-06-13 18:12:26 +00:00
Luka Govedič
f98548b9da
[torch.compile][ROCm] Fuse quantization onto attention using a torch.compile pass (#16756) 
Signed-off-by: Luka Govedič <lgovedic@redhat.com>
Co-authored-by: Sage Moore <sage@neuralmagic.com>
 2025-06-12 08:31:04 -07:00