Commits · f26dbd0960d6186d61c11954ba16329d7599fe08 · OpenDAS / dynamo · GitLab

15 Dec, 2025 6 commits
- fix: Fix kvbm nixl connector for vllm upgrade (#4957) · f26dbd09
  Kris Hung authored Dec 16, 2025
  
  f26dbd09
- chore: bump minijinja to 2.14.0 (#4949) · 24ac6c15
  KrishnanPrash authored Dec 15, 2025
```
Signed-off-by: Krishnan Prashanth <kprashanth@nvidia.com>
```
  24ac6c15
- feat: add useMocker field to DGDR (#4813) · 4e6493ae
  hhzhang16 authored Dec 15, 2025
```
Signed-off-by: Hannah Zhang <hannahz@nvidia.com>
```
  4e6493ae
- fix: reduce spurious resource updates (#4868) · c341ee8d
  Julien Mancuso authored Dec 15, 2025
```
Signed-off-by: Julien Mancuso <jmancuso@nvidia.com>
Co-authored-by: tmontfort <tmontfort@nvidia.com>
```
  c341ee8d
- docs: update support matrix with build dependency version history (#4953) · 8189e854
  dagil-nvidia authored Dec 15, 2025
```
Signed-off-by: Dan Gil <dagil@nvidia.com>
```
  8189e854
- feat: Mistral-3-large support (#4885) · 44a2cba9
  jthomson04 authored Dec 14, 2025
```
Signed-off-by: jthomson04 <jwillthomson19@gmail.com>
```
  44a2cba9
13 Dec, 2025 2 commits
- docs: router cost function mermaid (#4943) · 87112373
  Yan Ru Pei authored Dec 12, 2025
```
Signed-off-by: PeaBrane <yanrpei@gmail.com>
```
  87112373
- feat: Support for field include_stop_str_in_output (#4924) · 45e881d3
  KrishnanPrash authored Dec 12, 2025
```
Signed-off-by: Krishnan Prashanth <kprashanth@nvidia.com>
```
  45e881d3
12 Dec, 2025 10 commits
- feat: support decentralized router with NATs core (#4921) · ef3027bd
  Yan Ru Pei authored Dec 12, 2025
```
Signed-off-by: PeaBrane <yanrpei@gmail.com>
```
  ef3027bd
- feat: allow user to input gpu cost in profiler webui (#4935) · b24ccd29
  Hongkuan Zhou authored Dec 12, 2025
```
Signed-off-by: hongkuanz <hongkuanz@nvidia.com>
```
  b24ccd29
- chore: enable local indexers for sglang and trtllm (#4932) · c2a29f80
  Yan Ru Pei authored Dec 12, 2025
```
Signed-off-by: PeaBrane <yanrpei@gmail.com>
```
  c2a29f80
- fix: cleaner prefill + decode discovery (#4927) · feb49914
  Yan Ru Pei authored Dec 12, 2025
```
Signed-off-by: PeaBrane <yanrpei@gmail.com>
```
  feb49914
- fix: Initialize logger earlier to capture startup logs (#3570) · 0de75a87
  bin.pan authored Dec 13, 2025
```
Signed-off-by: bin <bin.pan@daocloud.io>
```
  0de75a87
- fix: health check race condition by waiting for watch stream discovery (#4876) · 32bfc0f3
  Tzu-Ling Kan authored Dec 11, 2025
```
Signed-off-by: tzulingk@nvidia.com <tzulingk@nvidia.com>
```
  32bfc0f3
- test: remove tcp test xfail for tcp after automatic port assignments (#4916) · 44f57a22
  Biswa Panda authored Dec 11, 2025
  
  44f57a22
- feat(operator): DGD service status replica information (#4863) · aa1bc3c5
  Thomas Montfort authored Dec 11, 2025
  
  aa1bc3c5
- fix: prefill discovery (#4911) · f790e921
  Yan Ru Pei authored Dec 11, 2025
```
Signed-off-by: PeaBrane <yanrpei@gmail.com>
```
  f790e921
- feat: DeepSeek V3.2 chat template support (#4797) · 1efc7d63
  Vladislav Nosivskoy authored Dec 12, 2025
```
Signed-off-by: Vladislav Nosivskoy <vladnosiv@gmail.com>
Co-authored-by: Ryan McCormick <rmccormick@nvidia.com>
```
  1efc7d63
11 Dec, 2025 22 commits
- feat: use os assigned free port for tcp rpc (#4891) · 64a30d3d
  Biswa Panda authored Dec 11, 2025
  
  64a30d3d
- feat: worker-local KvIndexer in KvEventPublisher (#4519) · 33249945
  Karen Chung authored Dec 11, 2025
```
Co-authored-by: Yan Ru Pei <yanrpei@gmail.com>
```
  33249945
- feat: early rejection based on active prefill tokens (#4837) · 10b01b45
  Yan Ru Pei authored Dec 11, 2025
```
Signed-off-by: PeaBrane <yanrpei@gmail.com>
```
  10b01b45
- ci: Update the pre_merge markers to exclude gpu_2 tests running (#4909) · 82577b06
  Pavithra Vijayakrishnan authored Dec 11, 2025
```
Signed-off-by: pvijayakrish <pvijayakrish@nvidia.com>
```
  82577b06
- test: remove pytest_runtestloop (#4886) · d0e95c39
  Yan Ru Pei authored Dec 11, 2025
```
Signed-off-by: PeaBrane <yanrpei@gmail.com>
```
  d0e95c39
- feat: install Run:ai model streamer for vllm (#4848) · 53cec4ac
  Guangtong Bai authored Dec 11, 2025
```
Signed-off-by: Guangtong Bai <guangtong.bai@gmail.com>
```
  53cec4ac
- fix: relax pydantic requirements (#4865) · 2ef0f5fb
  Dmitry Tokarev authored Dec 11, 2025
```
Signed-off-by: Dmitry Tokarev <dtokarev@nvidia.com>
```
  2ef0f5fb
- docs: update docs to remove KVBM cuda graph limitation (#4902) · 5c2415d8
  Kyle McGill authored Dec 11, 2025
  
  5c2415d8
- fix: remove small --gpu-memory-utilization to avoid OOM due to vllm upgrade (#4899) · 31f31e8e
  Ziqi Fan authored Dec 11, 2025
```
Signed-off-by: Ziqi Fan <ziqif@nvidia.com>
```
  31f31e8e
- test: graceful shutdown for tcp req plane (#4883) · 5e96c9a8
  Biswa Panda authored Dec 11, 2025
  
  5e96c9a8
- fix: [lora] refactor test and clean up examples (#4884) · 3c2b72b0
  Biswa Panda authored Dec 11, 2025
  
  3c2b72b0
- ci: Fixing sccache behavior (#4866) · 242a4d5b
  Dillon Cullinan authored Dec 11, 2025
```
Signed-off-by: Dillon Cullinan <dcullinan@nvidia.com>
Signed-off-by: Anant Sharma <anants@nvidia.com>
Co-authored-by: Anant Sharma <anants@nvidia.com>
```
  242a4d5b
- fix: Fix CUDA_VISIBLE_DEVICES for agg_multimodal_epd.sh (#4888) · 9b7152d7
  Kris Hung authored Dec 11, 2025
  
  9b7152d7
- test: add dynamic port allocation for fault_tolerant test execution (#4835) · c6b440e4
  Keiven C authored Dec 10, 2025
```
Signed-off-by: Keiven Chang <keivenchang@users.noreply.github.com>
Co-authored-by: Keiven Chang <keivenchang@users.noreply.github.com>
```
  c6b440e4
- refactor: simplify, remove --dev-image option from build.sh (#4867) · 111e08c5
  Keiven C authored Dec 10, 2025
```
Signed-off-by: Keiven Chang <keivenchang@users.noreply.github.com>
Co-authored-by: Keiven Chang <keivenchang@users.noreply.github.com>
```
  111e08c5
- feat: add cached tokens prometheus metric (#4534) · f8bb53c0
  Vladislav Nosivskoy authored Dec 11, 2025
```
Signed-off-by: Vladislav Nosivskoy <vladnosiv@gmail.com>
Co-authored-by: Ryan McCormick <rmccormick@nvidia.com>
```
  f8bb53c0
- feat: include ttft and total request time in nvext (engine-agnostic) (#4880) · 2845aa1f
  Yan Ru Pei authored Dec 10, 2025
```
Signed-off-by: PeaBrane <yanrpei@gmail.com>
Signed-off-by: Yan Ru Pei <yanrpei@gmail.com>
Co-authored-by: Keiven C <213854356+keivenchang@users.noreply.github.com>
```
  2845aa1f
- chore: TRTLLM 1.2.0rc4 (#4836) · 52503032
  Dmitry Tokarev authored Dec 10, 2025
```
Signed-off-by: Dmitry Tokarev <dtokarev@nvidia.com>
```
  52503032
- test: Ensure Request Cancellation, Migration, Rejection Work with TCP transport (#4875) · 00f8615e
  Jacky authored Dec 10, 2025
```
Signed-off-by: Jacky <18255193+kthui@users.noreply.github.com>
```
  00f8615e
- fix: worker to graceful shutdown after finishing in-flight requests (#4838) · c8845b41
  Hongkuan Zhou authored Dec 10, 2025
```
Signed-off-by: hongkuanz <hongkuanz@nvidia.com>
Co-authored-by: Biswa Panda <biswa.panda@gmail.com>
```
  c8845b41
- chore: add gb200 fp4 8k1k (#4874) · 69817c2d
  ishandhanani authored Dec 10, 2025
  
  69817c2d
- test: router + trtllm CI tests (#4805) · ac8d36c6
  Karen Chung authored Dec 10, 2025
  
  ac8d36c6