Tyler Michael Smith
9df42d2000
[Build] Update pre-commit pip-compile hook to use cu128 torch backend
...
Matches the test.txt lockfile change to CUDA 12.8.
Signed-off-by: Tyler Michael Smith <tlrmchlsmth@gmail.com >
2026-02-18 16:41:53 -05:00
Tyler Michael Smith and Claude Opus 4.6
8d151aa148
[Build] Recompile test.txt lockfile with cu128 torch backend
...
The lockfile was compiled with --torch-backend cu129, pinning
torch==2.10.0+cu129. This breaks Docker builds that use CUDA_VERSION=12.8
because the cu128 PyTorch index does not carry +cu129 wheels.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com >
Signed-off-by: Tyler Michael Smith <tlrmchlsmth@gmail.com >
2026-02-18 14:38:23 -05:00
Tyler Michael Smith and Claude Opus 4.6
11a13b0fd3
[Docker] Add BUILDER_CUDA_VERSION to decouple build and runtime CUDA versions
...
Allow compiling csrc/ and extensions (DeepGEMM, EP kernels) with a
different CUDA toolkit than the one shipped in the final runtime image.
BUILDER_CUDA_VERSION controls the devel base image used for compilation,
while CUDA_VERSION selects the runtime base image and PyTorch wheel index.
Override the CUDA_VERSION env var inherited from the nvidia base image in
the build stages so PyTorch index URLs resolve to the runtime version.
Update the CUDA 13.0 release pipeline entries to pass the new arg.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com >
Signed-off-by: Tyler Michael Smith <tlrmchlsmth@gmail.com >
2026-02-18 14:38:23 -05:00
Amr Mahdi and GitHub
df3f537a66
[CI] Remove unused precompiled wheel args from image build ( #34767 )
...
Signed-off-by: Amr Mahdi <amrmahdi@meta.com >
2026-02-17 18:58:18 -08:00
Matthew Bonanni and GitHub
7743152957
[Attention] Refactor check_and_update_config ( #33600 )
...
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com >
2026-02-17 17:06:54 -08:00
Wentao Ye and GitHub
ab33d2a629
[Feature] Decode Context Parallel support for GPU model runner v2 ( #34179 )
...
Signed-off-by: yewentao256 <zhyanwentao@126.com >
2026-02-17 16:27:15 -08:00
Woosuk Kwon and GitHub
be3af2d29e
[Model Runner V2] Further simplification for PP ( #34724 )
...
Signed-off-by: Woosuk Kwon <woosuk@inferact.ai >
2026-02-17 15:18:18 -08:00
c656ba3b4d
[Kernel] Triton-based Top-k and Top-p sampler kernels ( #33538 )
...
Signed-off-by: js_park <cakeng@naver.com >
Signed-off-by: Jongseok Park <37990712+cakeng@users.noreply.github.com >
Signed-off-by: Sunga Kim <sunga.kim@berkeley.edu >
Signed-off-by: Nick Hill <nickhill123@gmail.com >
Co-authored-by: Sunga Kim <sunga.kim@berkeley.edu >
Co-authored-by: Nick Hill <nickhill123@gmail.com >
2026-02-17 23:14:30 +00:00
dc5fa77a4e
[Bugfix][MTP][Sparse MLA] Allow sparse MLA with MTP to run with FULL cudagraphs ( #34457 )
...
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com >
Co-authored-by: Jee Jee Li <pandaleefree@gmail.com >
2026-02-17 14:01:27 -05:00
Flora Feng and GitHub
1e4a084c8e
[CI] Fix flaky test_parsable_context ( #34717 )
...
Signed-off-by: sfeng33 <4florafeng@gmail.com >
2026-02-17 18:42:52 +00:00
Richard Zou and GitHub
7967e854da
[BugFix] Fix sp tests ( #34716 )
...
Signed-off-by: Richard Zou <zou3519@gmail.com >
2026-02-17 17:07:56 +00:00
6bd6d0c3c1
Fixed whisper CPU test that does not spawn properly. ( #34324 )
...
Signed-off-by: Anna Mayne <anna.mayne@arm.com >
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk >
2026-02-17 06:46:23 -08:00
Nicolò Lucchesi and GitHub
8e962fef5f
[CI][Nixl] Add CrossLayer KV layout tests ( #34615 )
...
Signed-off-by: NickLucche <nlucches@redhat.com >
2026-02-17 21:35:40 +08:00
Cyrus Leung and GitHub
574fe75245
[Renderer] Move InputPreprocessor into Renderer (2/2) ( #34560 )
...
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk >
2026-02-17 05:29:01 -08:00
junuxyz and GitHub
c61a98f529
[CI][BugFix] ShellCheck cleanup to remove baseline and preserve runtime behavior ( #34514 )
...
Signed-off-by: junuxyz <216036880+junuxyz@users.noreply.github.com >
2026-02-17 12:22:56 +00:00
Harry Mellor and GitHub
28bffe9466
Fix docs build warning ( #34686 )
...
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com >
2026-02-17 02:31:40 -08:00
ad65177a19
[Bugfix] Fix 'remove_instance_endpoint' method logic in disagg_proxy_demo ( #32922 )
...
Signed-off-by: ChenqianCao <39755070+ChenqianCao@users.noreply.github.com >
Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com >
2026-02-17 10:06:53 +00:00
d44a5b6c47
Remove dead bitsandbytes CxB code from 8-bit inference path ( #34633 )
...
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com >
2026-02-17 01:49:14 -08:00
Jiangyun Zhu and GitHub
1d65283e95
Revert "[Models] Fuse Qwen3.5 GDN's qkvz_proj and ba_proj" ( #34683 )
2026-02-17 01:29:27 -08:00
c464b57374
[Ray] Propagate third-party env vars to Ray workers via prefix matching ( #34383 )
...
Signed-off-by: Kourosh Hakhamaneshi <kourosh@anyscale.com >
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-02-17 01:08:42 -08:00
Amr Mahdi and GitHub
c5c38e152a
[CI] Fix bake config artifact path for AMI rebuild pipeline ( #34656 )
...
Signed-off-by: Amr Mahdi <amrmahdi@meta.com >
2026-02-17 06:39:44 +00:00
Woosuk Kwon and GitHub
d00df624f3
[Model Runner V2] Minor refactoring for penalties ( #34662 )
...
Signed-off-by: Woosuk Kwon <woosuk@inferact.ai >
2026-02-16 21:43:00 -08:00
Woosuk Kwon and GitHub
9752da9d9c
[Model Runner V2] Minor simplification for BadWordsState ( #34669 )
...
Signed-off-by: Woosuk Kwon <woosuk@inferact.ai >
2026-02-16 21:27:24 -08:00
Woosuk Kwon and GitHub
04925b2202
[Model Runner V2] Minor cleanup for PP ( #34666 )
...
Signed-off-by: Woosuk Kwon <woosuk@inferact.ai >
2026-02-16 19:15:31 -08:00
Woosuk Kwon and GitHub
d74278fb67
[Model Runner V2] Fix unintended CPU-GPU sync in make_dummy ( #34667 )
...
Signed-off-by: Woosuk Kwon <woosuk@inferact.ai >
2026-02-16 19:00:29 -08:00
b68fd899d1
[Bugfix] Fix fused MoE int32 overflow in stride*offset without perf regression ( #34507 )
...
Signed-off-by: haosdent <haosdent@gmail.com >
Co-authored-by: Michael Goin <mgoin64@gmail.com >
2026-02-16 17:58:49 -08:00
Aneesh Puttur and GitHub
0b5f9b7204
[CI] Enable mypy import following for vllm/v1/kv_offload ( #34639 )
...
Signed-off-by: Aneesh Puttur <aneeshputtur@gmail.com >
2026-02-17 09:58:15 +08:00
zhanqiuhu and GitHub
9a8853f781
[Core] Pipeline Parallel support for Model Runner V2 ( #33960 )
...
Signed-off-by: Zhanqiu Hu <zh338@cornell.edu >
2026-02-16 17:48:16 -08:00
387a1898d9
[Model Runner V2] support bad_words sampling param ( #33433 )
...
Signed-off-by: zhuhaoran <zhuhaoran.zhr@alibaba-inc.com >
Signed-off-by: Woosuk Kwon <woosuk@inferact.ai >
Co-authored-by: Woosuk Kwon <woosuk@inferact.ai >
2026-02-16 16:36:06 -08:00
roikoren755 and GitHub
3b30e61507
[NemotronH] Do not force router to run in fp32 ( #34582 )
...
Signed-off-by: Roi Koren <roik@nvidia.com >
2026-02-16 10:15:32 -08:00
Alexei-V-Ivanov-AMD and GitHub
824f9e8f3c
Targeting the MI355 agent pool with all existing tests ( #34629 )
...
Signed-off-by: Alexei V. Ivanov <alexei.ivanov@amd.com >
2026-02-16 17:02:27 +00:00
Nicolò Lucchesi and GitHub
6cc403e67d
[Bugfix][CI] Fix flaky entrypoints/openai/test_response_api_with_harmony.py::test_function_calling[openai/gpt-oss-20b] ( #34624 )
...
Signed-off-by: NickLucche <nlucches@redhat.com >
2026-02-16 16:11:07 +00:00
Almog Tavor and GitHub
72d5951d02
[Bugfix] Treat generation_config max_tokens as default not ceiling ( #34063 )
...
Signed-off-by: almogtavor <almogtavor@gmail.com >
2026-02-16 07:58:24 -08:00
a3205beffb
[CI] Enable mypy coverage for individual excluded files ( #34292 )
...
Signed-off-by: Lucas Kabela <lucaskabela@meta.com >
Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com >
2026-02-16 07:34:29 -08:00
Christian Pinto and GitHub
6930becd45
(bugfix): Fixed encode in LLM entrypoint for IOProcessr plugin prompts ( #34618 )
...
Signed-off-by: Christian Pinto <christian.pinto@ibm.com >
2026-02-16 07:33:55 -08:00
Andreas Karatzas and GitHub
03a8770a6d
[ROCm][CI] Fix plugins test group; updating terratorch and dependencies ( #34589 )
...
Signed-off-by: Andreas Karatzas <akaratza@amd.com >
2026-02-16 07:33:42 -08:00
Yiqi Xue and GitHub
bc56a1d56e
[Bugfix] Fix ARC touch KeyError for non-ready T1 blocks in kv offload ( #34576 )
...
Signed-off-by: Yiqi Xue <xuey666@gmail.com >
2026-02-16 07:33:19 -08:00
danisereb and GitHub
ec7d9e6745
Fix call to moe_mk in modelopt MoE modules (required for LoRA) ( #34575 )
...
Signed-off-by: Daniel Serebrenik <daserebrenik@nvidia.com >
2026-02-16 07:33:09 -08:00
Isotr0py and GitHub
3bb4e4311c
[Models] Fuse Qwen3.5 GDN's qkvz_proj and ba_proj ( #34492 )
...
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn >
2026-02-16 07:32:51 -08:00
Amr Mahdi and GitHub
08f8c198ae
[CI] Disable precompiled wheel path in CI image builds ( #34606 )
...
Signed-off-by: Amr Mahdi <amrmahdi@meta.com >
2026-02-16 15:14:43 +00:00
Harry Mellor and GitHub
a21cedf4ff
Bump lm-eval version for Transformers v5 compatibility ( #33994 )
...
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com >
2026-02-16 05:24:35 -08:00
emricksini-h and GitHub
3ef74cde5d
[CI][Tracing] Fix race condition by adding server readiness check ( #34364 )
...
Attempt to resolve #34284 : "Metrics Tracing (2GPU)" fails with a
segmentation fault.
Signed-off-by: emricksini-h <emrick.birivoutin@hcompany.ai >
2026-02-16 12:57:39 +00:00
cd81cdb399
[Scheduler][ASR] Fix CrossAttn blocks per-request for Variable length encoder inputs ( #31058 )
...
Signed-off-by: Ekagra Ranjan <3116519+ekagra-ranjan@users.noreply.github.com >
Co-authored-by: Nicolò Lucchesi <nlucches@redhat.com >
2026-02-16 11:08:44 +00:00
Andreas Karatzas and GitHub
1e828573b4
[CI][Metrics] Stabilize tests with polling and subprocess guards ( #34566 )
...
test_abort_metrics_reset is flaky due to hardware-dependent
fixed sleeps: replace fixed sleeps with polling.
test_metrics_exist_run_batch passes even when the engine crashes
on startup (false positive): add subprocess lifecycle guards.
Signed-off-by: Andreas Karatzas <akaratza@amd.com >
2026-02-16 10:52:02 +00:00
a5ccc85c8c
[Bugfix] Fix Dynamo unexpected keyword argument ( #34320 )
...
Signed-off-by: Samu Tamminen <stammine@amd.com >
Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com >
2026-02-16 01:32:30 -08:00
Roger Wang and GitHub
b5475d0534
Revert "[Misc] fix qwen3.5 config" ( #34610 )
2026-02-16 01:06:05 -08:00
JJJYmmm and GitHub
9521002f0a
[Misc] fix qwen3.5 config ( #34604 )
2026-02-16 00:25:38 -08:00
Cyrus Leung and GitHub
ec17bdd894
[Renderer] Move InputPreprocessor into Renderer (1.5/2) ( #34598 )
...
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk >
2026-02-15 23:46:33 -08:00
Amr Mahdi and GitHub
bb59c90248
[CI] Write bake config to temp directory instead of repo root ( #34569 )
...
Signed-off-by: Amr Mahdi <amrmahdi@meta.com >
2026-02-15 22:15:47 -08:00
bnellnm and GitHub
5bff999d12
[Bugfix] Add method to swap quant_method on FusedMoE to fix LoRA issues ( #34453 )
...
Signed-off-by: Bill Nell <bnell@redhat.com >
2026-02-15 20:10:50 -08:00
Tyler Michael Smith