Commits · 83a3fe4e938b588ad0f51388d803aef49bcf2ccc · OpenDAS / dynamo · GitLab

05 Nov, 2025 2 commits
- chore: bump ai-configurator version in container to make gradio install optional (#4106) · 83a3fe4e
  Tushar Sharma authored Nov 04, 2025
```
Signed-off-by: Tushar Sharma <tusharma@nvidia.com>
```
  83a3fe4e
- feat: add --runtime-check flag to sanity_check.py (#4102) · 25652437
  Keiven C authored Nov 04, 2025
```
Signed-off-by: Keiven Chang <keivenchang@users.noreply.github.com>
Co-authored-by: Keiven Chang <keivenchang@users.noreply.github.com>
```
  25652437
04 Nov, 2025 16 commits
- fix: fix pod affinity in vllm recipe (#4107) · 5128e2b4
  Julien Mancuso authored Nov 04, 2025
```
Signed-off-by: Julien Mancuso <jmancuso@nvidia.com>
```
  5128e2b4
- feat: Image decoder in the frontend (#3971) · ae4b08ac
  milesial authored Nov 04, 2025
```
Signed-off-by: Alexandre Milesi <milesial@users.noreply.github.com>
```
  ae4b08ac
- fix: use mc backend for disagg (#4095) · 72b0aec1
  ishandhanani authored Nov 04, 2025
  
  72b0aec1
- fix: oss-gpt-120b model path in recipe (#4085) · a2e748ac
  Biswa Panda authored Nov 04, 2025
  
  a2e748ac
- feat: use configmap instead of PVC for profiling results (#3981) · d3706221
  Hongkuan Zhou authored Nov 04, 2025
```
Signed-off-by: hongkuanz <hongkuanz@nvidia.com>
Signed-off-by: Hannah Zhang <hannahz@nvidia.com>
Co-authored-by: Hannah Zhang <hannahz@nvidia.com>
```
  d3706221
- docs: update wideep documentation to precompile DeepGemm kernels beforehand (#4096) · 51c103b2
  Tushar Sharma authored Nov 04, 2025
```
Signed-off-by: Tushar Sharma <tusharma@nvidia.com>
Co-authored-by: coderabbitai[bot] <136622811+coderabbitai[bot]@users.noreply.github.com>
```
  51c103b2
- fix: add prefill metrics support for TensorRT-LLM disaggregated mode (#3983) · ba51c683
  Keiven C authored Nov 04, 2025
```
Signed-off-by: Keiven Chang <keivenchang@users.noreply.github.com>
Co-authored-by: Keiven Chang <keivenchang@users.noreply.github.com>
```
  ba51c683
- fix: add prefill metrics support for SGLang disaggregated mode (#3992) · 49ecfe60
  Keiven C authored Nov 04, 2025
```
Signed-off-by: Keiven Chang <keivenchang@users.noreply.github.com>
```
  49ecfe60
- fix: test_json_log_capture (#4074) · c9ca5c40
  Thomas Montfort authored Nov 04, 2025
```
Co-authored-by: mohammedabdulwahhab <furkhan324@berkeley.edu>
```
  c9ca5c40
- fix: add PodClique watch to trigger DGD reconciliation (#4076) · 816a4c64
  Julien Mancuso authored Nov 04, 2025
```
Signed-off-by: Julien Mancuso <jmancuso@nvidia.com>
```
  816a4c64
- feat: kvbm v2 memory crate (#4012) · 3659c82e
  Ryan Olson authored Nov 04, 2025
```
Signed-off-by: Ryan Olson <rolson@nvidia.com>
```
  3659c82e
- feat(sglang): Add dummy warmup req for prefill (#4058) · bc02088e
  YAMY authored Nov 03, 2025
```
Co-authored-by: ishandhanani <82981111+ishandhanani@users.noreply.github.com>
```
  bc02088e
- fix: add missing close brace in dynamo.code-workspace (#4055) · 4beada35
  Lin Yiming authored Nov 04, 2025
```
Signed-off-by: Lin Yiming <roymatrix.lin@gmail.com>
```
  4beada35
- fix: update router standalone to use updated vLLM API (#4079) · 81b8af8d
  Sean SH Choi authored Nov 03, 2025
```
Signed-off-by: Sean Choi <sechoi@nvidia.com>
```
  81b8af8d
- docs: Cleanup KServe GRPC docs (#4081) · 98842a03
  Ryan McCormick authored Nov 03, 2025
  
  98842a03
- docs: add KServe gRPC frontend docs (#4020) · ed0849b2
  GuanLuo authored Nov 04, 2025
  
  ed0849b2
03 Nov, 2025 10 commits
- fix: add prefill load balance method args for deepseek-r1 (#4051) · 04a6eb6c
  Biswa Panda authored Nov 03, 2025
```
Co-authored-by: dagil-nvidia <dagil@nvidia.com>
```
  04a6eb6c
- feat: Reject unsupported parameters with 400 Bad Request (#4021) · c837b5ba
  KrishnanPrash authored Nov 03, 2025
```
Signed-off-by: Krishnan Prashanth <kprashanth@nvidia.com>
```
  c837b5ba
- feat: grab num_experts info from model info if possible (#4060) · f31a1dad
  hhzhang16 authored Nov 03, 2025
```
Signed-off-by: Hannah Zhang <hannahz@nvidia.com>
Signed-off-by: hongkuanz <hongkuanz@nvidia.com>
Co-authored-by: hongkuanz <hongkuanz@nvidia.com>
Co-authored-by: Hongkuan Zhou <tedzhouhk@gmail.com>
```
  f31a1dad
- chore: bump fastapi version to 0.120.1 (#4002) · 92909c8a
  Tushar Sharma authored Nov 03, 2025
```
Signed-off-by: Tushar Sharma <tusharma@nvidia.com>
```
  92909c8a
- fix: vllm disagg launch script should not enable kv routing (#4063) · 2d7f5fc5
  Yan Ru Pei authored Nov 03, 2025
```
Signed-off-by: PeaBrane <yanrpei@gmail.com>
```
  2d7f5fc5
- chore: relax DynamoStatLoggerPublisher.record() signature (#3986) · 3555b72f
  GuanLuo authored Nov 04, 2025
```
Signed-off-by: Guan Luo <41310872+GuanLuo@users.noreply.github.com>
```
  3555b72f
- chore: disagg + router k8s recipe minor changes (#4050) · ba99d24a
  Yan Ru Pei authored Nov 03, 2025
```
Signed-off-by: PeaBrane <yanrpei@gmail.com>
```
  ba99d24a
- ci: Fix frontend pod detection in deploy tests (#4054) · e3ba40e8
  Dillon Cullinan authored Nov 03, 2025
```
Signed-off-by: Dillon Cullinan <dcullinan@nvidia.com>
```
  e3ba40e8
- chore: Remove old DisaggregatedRouter, making etcd presence optional (#4011) · dadf0e22
  Graham King authored Nov 03, 2025
```
Signed-off-by: Graham King <grahamk@nvidia.com>
```
  dadf0e22
- test: move backend unit tests (#3958) · 30d85883
  Pavithra Vijayakrishnan authored Nov 02, 2025
```
Signed-off-by: pvijayakrish <pvijayakrish@nvidia.com>
```
  30d85883
01 Nov, 2025 4 commits
- ci: OPS-1745: Deploy operator once for deploy tests (#4022) · 9d765839
  Dillon Cullinan authored Nov 01, 2025
```
Signed-off-by: Dillon Cullinan <dcullinan@nvidia.com>
Signed-off-by: Dillon Cullinan <dcullinan92@gmail.com>
```
  9d765839
- feat: restructure dsr1 recipes and add gb200 (#3891) · b73e6eb5
  Rohan Varma authored Oct 31, 2025
```
Signed-off-by: Rohan Varma <rohanv@nvidia.com>
```
  b73e6eb5
- fix: small fixes for OTEL trace implementation (#3734) · e20adb44
  Thomas Montfort authored Oct 31, 2025
  
  e20adb44
- feat: efficient serving of multiple mockers (#3997) · 6ff49edb
  Yan Ru Pei authored Oct 31, 2025
```
Signed-off-by: PeaBrane <yanrpei@gmail.com>
```
  6ff49edb
31 Oct, 2025 8 commits
- fix: use full path for recipes directory (#3989) · 17139df9
  Anant Sharma authored Oct 31, 2025
```
Signed-off-by: Anant Sharma <anants@nvidia.com>
```
  17139df9
- fix: reduce max ISL when benchmarking prefill (#4030) · f2414f51
  Hongkuan Zhou authored Oct 31, 2025
```
Signed-off-by: hongkuanz <hongkuanz@nvidia.com>
```
  f2414f51
- chore: KVBM pip wheel (#3826) · 6afa679c
  Richard Huo authored Oct 31, 2025
```
Signed-off-by: Anant Sharma <anants@nvidia.com>
Co-authored-by: Anant Sharma <anants@nvidia.com>
```
  6afa679c
- fix: remove metric publisher tight loop (#4028) · e5c109d8
  Yan Ru Pei authored Oct 31, 2025
```
Signed-off-by: PeaBrane <yanrpei@gmail.com>
```
  e5c109d8
- fix: remove quote (#4027) · 89bb3fe9
  Alec authored Oct 31, 2025
```
Signed-off-by: alec-flowers <aflowers@nvidia.com>
```
  89bb3fe9
- fix: move new files from components to examples backend (#4023) · e6809ff0
  Anant Sharma authored Oct 31, 2025
```
Signed-off-by: Anant Sharma <anants@nvidia.com>
```
  e6809ff0
- fix: test uploading to not overwrite itself for pytest trtllm, sglang, and vllm jobs (#3859) · e4faec9f
  Nate Mailhot authored Oct 31, 2025
  
  e4faec9f
- 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