Commits · 8679faa35d2ed872babd1f8d160f5f259e01d1aa · OpenDAS / dynamo · GitLab

21 Apr, 2026 8 commits
- test(sglang): add pre_merge diffusion CI smoke tests for T2I and T2V (#8409) · 8679faa3
  Krishnan Prashanth authored Apr 21, 2026
```
Signed-off-by: Krishnan Prashanth <kprashanth@nvidia.com>
```
  8679faa3
- docs(operator): clarify Interconnect and RDMA field semantics in HardwareSpec (#8300) · 19ecf46f
  Dr. Stefan Schimanski authored Apr 21, 2026
```
Signed-off-by: Dr. Stefan Schimanski <sschimanski@nvidia.com>
```
  19ecf46f
- fix(vllm): include image geometry in multimodal hash preimage (#8341) · 5103efdb
  Neelay Shah authored Apr 20, 2026
```
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
```
  5103efdb
- fix(snapshot): replace PID-1 SIGUSR1/SIGCONT contract with file sentinels (#8403) · ca15a79d
  Schwinn Saereesitthipitak authored Apr 20, 2026
```
Signed-off-by: Schwinn Saereesitthipitak <schwinns@nvidia.com>
```
  ca15a79d
- fix: Validate multimodal media URLs and load paths (#8282) · 8fba4f56
  zhongdaor-nv authored Apr 20, 2026
```
Signed-off-by: zhongdaor <zhongdaor@nvidia.com>
```
  8fba4f56
- chore(mocker): panic loudly on scheduler invariant violations (#8406) · f701319e
  Yan Ru Pei authored Apr 20, 2026
```
Signed-off-by: PeaBrane <yanrpei@gmail.com>
```
  f701319e
- docs(recipes): clarify Qwen3-235B DEEPGEMM requires Blackwell (SM100+) (#8410) · 781e4100
  Yuewei Na authored Apr 20, 2026
```
Signed-off-by: Yuewei Na <nv-yna@users.noreply.github.com>
Co-authored-by: Yuewei Na <nv-yna@users.noreply.github.com>
```
  781e4100
- fix(planner/tests): scope test_advisory_mode stubs so they don't leak (#8418) · 508aed84
  Dmitry Tokarev authored Apr 20, 2026
```
Signed-off-by: Dmitry Tokarev <dtokarev@nvidia.com>
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
```
  508aed84
20 Apr, 2026 28 commits
- chore(deps): bump vLLM to 0.19.1 (#8370) · d59b56be
  Ayush Agarwal authored Apr 20, 2026
```
Signed-off-by: ayushag <ayushag@nvidia.com>
```
  d59b56be
- fix(parsers): recover Kimi K2 tool calls when section_end is missing (#8208) · 47cfbd46
  Keiven C authored Apr 20, 2026
```
Signed-off-by: Keiven Chang <keivenchang@users.noreply.github.com>
```
  47cfbd46
- feat: (Router) Guard threshold-based rejection logic when CLI thresholds are unset (#8333) · adfc02d5
  Karen Chung authored Apr 20, 2026
  
  adfc02d5
- fix: update the tool calling functionalities for sglang frontend processor to... · 8fd7de9a
  Richard Huo authored Apr 20, 2026
```
fix: update the tool calling functionalities for sglang frontend processor to match with the latest sglang implementation (#8269)
```
  8fd7de9a
- fix(operator/deploy/tilt): build helm dependencies automatically (#8155) · 01688850
  Dr. Stefan Schimanski authored Apr 20, 2026
```
Signed-off-by: Dr. Stefan Schimanski <sschimanski@nvidia.com>
```
  01688850
- fix(container): preserve file permissions in SGLang dev venv setup (#8408) · 303520f8
  Krishnan Prashanth authored Apr 20, 2026
```
Signed-off-by: Krishnan Prashanth <kprashanth@nvidia.com>
```
  303520f8
- fix(kv-router): allow unit block size in slot tracking (#8395) · cdb40ec6
  Yan Ru Pei authored Apr 20, 2026
```
Signed-off-by: PeaBrane <yanrpei@gmail.com>
```
  cdb40ec6
- fix: llava EPD deployment. add llava coverage (#8330) · e71f1d2b
  GuanLuo authored Apr 20, 2026
```
Signed-off-by: Guan Luo <41310872+GuanLuo@users.noreply.github.com>
Signed-off-by: GuanLuo <41310872+GuanLuo@users.noreply.github.com>
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
```
  e71f1d2b
- fix(operator): reconcile DynamoGraphDeployment on PodCliqueScalingGroup status changes (#8328) · c68d159f
  Julien Mancuso authored Apr 20, 2026
  
  c68d159f
- fix(planner): backfill max_num_batched_tokens from discovery for VirtualConnector (#8042) · 44190094
  jthomson04 authored Apr 20, 2026
```
Signed-off-by: jthomson04 <jwillthomson19@gmail.com>
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
```
  44190094
- docs(mocker): clarify mocker image placeholder (#8274) · 5c8f8ffb
  Ryan McCormick authored Apr 20, 2026
  
  5c8f8ffb
- fix(protocols): always serialize chat message content (null when absent) (#8372) · 846ce71a
  Ryan McCormick authored Apr 20, 2026
```
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
```
  846ce71a
- fix: Add accelerate to sglang runtime image for diffusion workers (#8357) · 7b979c00
  Krishnan Prashanth authored Apr 20, 2026
```
Signed-off-by: Krishnan Prashanth <kprashanth@nvidia.com>
```
  7b979c00
- ci: expose cpu_parallel_mode and run planner tests with -n 0 (#8378) · 2e3d4a92
  Dmitry Tokarev authored Apr 20, 2026
```
Signed-off-by: Dmitry Tokarev <dtokarev@nvidia.com>
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
```
  2e3d4a92
- docs(vllm-omni): note ARM64 not supported; flag GLM-Image as experimental (#8373) · 8291bfcc
  Ayush Agarwal authored Apr 20, 2026
```
Signed-off-by: ayushag <ayushag@nvidia.com>
```
  8291bfcc
- fix(planner): normalize model_name case in KubernetesConnector comparisons (#8384) · 073da7b1
  Hongkuan Zhou authored Apr 20, 2026
```
Signed-off-by: hongkuanz <hongkuanz@nvidia.com>
Co-authored-by: brluobt <brluo@nvidia.com>
```
  073da7b1
- fix(operator): emit Kubernetes syntax from LWS deployer so kubelet expands... · da7d6724
  Julien Mancuso authored Apr 20, 2026
```
fix(operator): emit Kubernetes  syntax from LWS deployer so kubelet expands the leader hostname in direct-python container args (#8369)
```
  da7d6724
- docs: add --prerelease=allow to sglang install commands (#8358) · 9054c882
  Krishnan Prashanth authored Apr 20, 2026
```
Signed-off-by: Krishnan Prashanth <kprashanth@nvidia.com>
```
  9054c882
- feat(planner): own AIC interpolation; fix MoE-DEP bugs in rapid mode (#8335) · 26188122
  Hongkuan Zhou authored Apr 20, 2026
```
Signed-off-by: hongkuanz <hongkuanz@nvidia.com>
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
```
  26188122
- fix(responses): own input chain in dynamo-protocols to accept Codex/Agents-SDK shapes (#8275) · 4ef28940
  ishandhanani authored Apr 20, 2026
  
  4ef28940
- feat(planner/replay): KV reuse awareness in load + throughput scaling (#8314) · c388483a
  Hongkuan Zhou authored Apr 20, 2026
```
Signed-off-by: hongkuanz <hongkuanz@nvidia.com>
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
```
  c388483a
- ci: set skopeo runner to be prod-skopeo-v1 (#8347) · db124db0
  Ran Rubin authored Apr 20, 2026
```
Signed-off-by: rrubin <rrubin@nvidia.com>
```
  db124db0
- feat(frontend): gate nvext response metadata behind extra_fields (#8252) · f437c8cf
  Biswa Panda authored Apr 20, 2026
```
Co-authored-by: AmeenP <ameenp360@gmail.com>
```
  f437c8cf
- feat: add staged frontend gauges (#8162) · 0c98d9a1
  Jie Hao authored Apr 20, 2026
  
  0c98d9a1
- docs: refresh quickstart to v1.0.1 and document partial v1.1.0-dev releases (#8364) · 7f58231b
  Dan Gil authored Apr 20, 2026
```
Signed-off-by: Dan Gil <dagil@nvidia.com>
```
  7f58231b
- feat: vLLM multinode elastic EP scaling support test (#8183) · f460f4e2
  Tzu-Ling Kan authored Apr 20, 2026
```
Signed-off-by: Tzu-Ling <tzulingk@nvidia.com>
```
  f460f4e2
- fix(sglang): add enable_trace to diffusion worker ServerArgs stub (#8332) · 1d154dcb
  Krishnan Prashanth authored Apr 20, 2026
```
Signed-off-by: Krishnan Prashanth <kprashanth@nvidia.com>
```
  1d154dcb
- fix: propagate vLLM --stream-interval to Dynamo frontend (#8101) · 714b4234
  Roger Wang authored Apr 19, 2026
```
Signed-off-by: Roger Wang <hey@rogerw.io>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
```
  714b4234
17 Apr, 2026 4 commits
- chore(kv-router): make sequences stop doing token math (#8260) · d94b350d
  Yan Ru Pei authored Apr 17, 2026
```
Signed-off-by: PeaBrane <yanrpei@gmail.com>
```
  d94b350d
- fix(health): use local endpoint registry for canary health checks (#8294) · c2ec3359
  Thomas Montfort authored Apr 17, 2026
```
Signed-off-by: Thomas Montfort <tmontfort@nvidia.com>
Signed-off-by: Thomas Montfort <tmontfort@nvidia.com>
Signed-off-by: tmontfort <tmontfort@nvidia.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
```
  c2ec3359
- fix(snapshot): resolve DRA GPU UUIDs from claims (#8292) · 2cc6d1e2
  Schwinn Saereesitthipitak authored Apr 17, 2026
  
  2cc6d1e2
- refactor(ci): switch sccache auth to IRSA web identity (#8324) · 8428c65f
  Saravana Periyasamy authored Apr 17, 2026
  
  8428c65f