Commits · 677e9b7f105f5a90b97fca80ef25d3ad0520e5f9 · OpenDAS / dynamo · GitLab

31 Oct, 2025 8 commits
- feat: allow for cluster-wide and namespace restricted operators in same cluster (#3966) · 677e9b7f
  Julien Mancuso authored Oct 31, 2025
```
Signed-off-by: Julien Mancuso <jmancuso@nvidia.com>
```
  677e9b7f
- feat: Media HTTP fetching and b64 decoding (#3967) · e30a3054
  milesial authored Oct 31, 2025
```
Signed-off-by: Alexandre Milesi <milesial@users.noreply.github.com>
```
  e30a3054
- feat: exclude k8 FT tests that need custom build from default run (#4017) · 20d1eb2e
  Tzu-Ling Kan authored Oct 31, 2025
```
Signed-off-by: tzulingk@nvidia.com <tzulingk@nvidia.com>
```
  20d1eb2e
- refactor: move backend deploy, launch and slurm files from components to examples (#3849) · 8bd37c96
  Anant Sharma authored Oct 31, 2025
```
Signed-off-by: Anant Sharma <anants@nvidia.com>
```
  8bd37c96
- ci: OPS-1744: Remove tool installs and test new CI image (#3991) · 78359046
  Dillon Cullinan authored Oct 31, 2025
```
Signed-off-by: Dillon Cullinan <dcullinan@nvidia.com>
```
  78359046
- fix: update model express common to use HF_HOME (if specified) (#3638) · 7f0ed081
  Biswa Panda authored Oct 31, 2025
  
  7f0ed081
- feat: Add prompt > seq_len k8 tests. (#3930) · c4abe9bc
  Tzu-Ling Kan authored Oct 30, 2025
```
Signed-off-by: tzulingk@nvidia.com <tzulingk@nvidia.com>
```
  c4abe9bc
- fix: introduce service discovery interface (1/n) (#3937) · 94ad5081
  mohammedabdulwahhab authored Oct 30, 2025
```
Signed-off-by: mohammedabdulwahhab <furkhan324@berkeley.edu>
```
  94ad5081
30 Oct, 2025 15 commits
- fix: platform helm chart value etcd preUpgrade not working (#3974) · bfb2574f
  Thomas Montfort authored Oct 30, 2025
  
  bfb2574f
- fix: Skip kvbm consolidator test if vllm is not installed (#4007) · 830ed660
  Kris Hung authored Oct 30, 2025
  
  830ed660
- feat: positional encoded sequence hashes (#4000) · e5ae505b
  Ryan Olson authored Oct 30, 2025
```
Signed-off-by: Ryan Olson <rolson@nvidia.com>
```
  e5ae505b
- feat: remove cluster wide logic from namespace restricted operator (#3934) · 6fc4c595
  hhzhang16 authored Oct 30, 2025
```
Signed-off-by: Hannah Zhang <hannahz@nvidia.com>
```
  6fc4c595
- chore: support for agg llama4 mulimodal (#3984) · 22d910a5
  Ayush Agarwal authored Oct 30, 2025
```
Signed-off-by: ayushag <ayushag@nvidia.com>
```
  22d910a5
- docs: Adding redirects to correct PR 3658 moves (#3985) · f2a3c638
  Andrew Schilling authored Oct 30, 2025
```
Signed-off-by: Andrew Schilling <aschilling@nvidia.com>
Co-authored-by: Kristen Kelleher <kkelleher@nvidia.com>
```
  f2a3c638
- chore: Revert "bump sglang version for 0.6.1 dynamo release (#3911)" (#3994) · 5f9d2025
  Tushar Sharma authored Oct 30, 2025
  
  5f9d2025
- chore: override docker build defaults for vllm + arm64 (#3980) · 15663b85
  Karen Chung authored Oct 30, 2025
  
  15663b85
- ci: Use nvcr.io image for sglang base container (#3996) · 7ca658f4
  Dillon Cullinan authored Oct 30, 2025
```
Signed-off-by: Dillon Cullinan <dcullinan@nvidia.com>
Signed-off-by: Tushar Sharma <tusharma@nvidia.com>
Co-authored-by: Tushar Sharma <tusharma@nvidia.com>
```
  7ca658f4
- chore: KV Router start using Store instead of etcd (#3945) · 13ce09ea
  Graham King authored Oct 30, 2025
```
Signed-off-by: Graham King <grahamk@nvidia.com>
```
  13ce09ea
- feat: allow Triton model config specification in TensorModelConfig (#3874) · eb3a486d
  GuanLuo authored Oct 31, 2025
```
Signed-off-by: Guan Luo <41310872+GuanLuo@users.noreply.github.com>
Signed-off-by: GuanLuo <41310872+GuanLuo@users.noreply.github.com>
```
  eb3a486d
- chore: bump nixl verison to 0.7.0 (#3936) · 1da9d70a
  Tushar Sharma authored Oct 30, 2025
```
Signed-off-by: Tushar Sharma <tusharma@nvidia.com>
```
  1da9d70a
- feat: Add KV event consolidator for KVBM (vllm) and router integration (#3725) · 18986010
  Kris Hung authored Oct 30, 2025
```
Signed-off-by: krishung5 <krish@nvidia.com>
```
  18986010
- chore: Improve KVBM test handling and update gitignore (#3987) · 95214e8b
  Kris Hung authored Oct 30, 2025
  
  95214e8b
- fix: disable lychee link checker until fixed (#3979) · 6bec8b31
  Nate Mailhot authored Oct 29, 2025
  
  6bec8b31
29 Oct, 2025 17 commits
- fix: NIXL connect header strip (#3959) · da10299c
  milesial authored Oct 29, 2025
```
Signed-off-by: Alexandre Milesi <milesial@users.noreply.github.com>
```
  da10299c
- fix: Operator defaults for liveness and readiness probes (#3968) · 62de76b6
  Thomas Montfort authored Oct 29, 2025
  
  62de76b6
- docs: Adding redirects (#3965) · 7b709c41
  Andrew Schilling authored Oct 29, 2025
```
Signed-off-by: Andrew Schilling <aschilling@nvidia.com>
```
  7b709c41
- fix: adjust kgateway setup (#3963) · d9e8d334
  atchernych authored Oct 29, 2025
```
Signed-off-by: Anna Tchernych <atchernych@nvidia.com>
```
  d9e8d334
- chore: add usage field to non-streaming responses by default (#3922) · 3fe5653b
  Ayush Agarwal authored Oct 29, 2025
```
Signed-off-by: ayushag <ayushag@nvidia.com>
Co-authored-by: Ryan McCormick <rmccormick@nvidia.com>
```
  3fe5653b
- chore: move jail stream readme (#3509) · 502df72b
  Ayush Agarwal authored Oct 29, 2025
```
Signed-off-by: ayushag <ayushag@nvidia.com>
Co-authored-by: Ryan McCormick <rmccormick@nvidia.com>
```
  502df72b
- chore: bump sglang version for 0.6.1 dynamo release (#3911) · 666e4aa7
  Tushar Sharma authored Oct 29, 2025
```
Signed-off-by: Tushar Sharma <tusharma@nvidia.com>
```
  666e4aa7
- feat: refactor vllm multimodal example (#3634) · 52090e2e
  Ayush Agarwal authored Oct 29, 2025
```
Signed-off-by: ayushag <ayushag@nvidia.com>
```
  52090e2e
- fix: autodetect Volcano installation (#3956) · 7d78fdad
  Julien Mancuso authored Oct 29, 2025
```
Signed-off-by: Julien Mancuso <jmancuso@nvidia.com>
```
  7d78fdad
- docs: add sphinx reredirects extension (#3962) · ce576c0c
  Anant Sharma authored Oct 29, 2025
```
Signed-off-by: Anant Sharma <anants@nvidia.com>
```
  ce576c0c
- fix: Refactored build.sh to allow users to install their local trtllm wheel (#3769) · 3077e1ec
  Kyle McGill authored Oct 29, 2025
```
Signed-off-by: Kyle McGill <101670481+nv-kmcgill53@users.noreply.github.com>
Co-authored-by: Anant Sharma <anants@nvidia.com>
```
  3077e1ec
- feat: add prefill worker metrics support for vLLM (#3949) · 4bce7f75
  Keiven C authored Oct 29, 2025
```
Signed-off-by: Keiven Chang <keivenchang@users.noreply.github.com>
```
  4bce7f75
- ci: fix variable processing into internal pipeline (#3946) · 18ade7d8
  Tushar Sharma authored Oct 29, 2025
```
Signed-off-by: Tushar Sharma <tusharma@nvidia.com>
Co-authored-by: Anant Sharma <anants@nvidia.com>
```
  18ade7d8
- fix(dynamo-run): Fix naming the model in single-process mode (#3955) · efa647b7
  Graham King authored Oct 29, 2025
```
Signed-off-by: Graham King <grahamk@nvidia.com>
```
  efa647b7
- fix: dynamo-run model name should default to remote path i.e. HFID (#3951) · a430bbb6
  Yan Ru Pei authored Oct 29, 2025
```
Signed-off-by: PeaBrane <yanrpei@gmail.com>
```
  a430bbb6
- fix: sanity_check.py 1) sglang Python site-packages check 2) adding... · f1552864
  Keiven C authored Oct 28, 2025
```
fix: sanity_check.py 1) sglang Python site-packages check 2) adding HuggingFace cache checking (#3890)
Signed-off-by: Keiven Chang <keivenchang@users.noreply.github.com>
```
  f1552864
- fix: profiler sidecar and other SLA-driven autodeployment fixes (#3932) · a7b703bd
  hhzhang16 authored Oct 28, 2025
```
Signed-off-by: Hannah Zhang <hannahz@nvidia.com>
```
  a7b703bd