Commits · 13058cbdea572690a50b5a92e8a4c9d1885801fe · OpenDAS / dynamo

"benchmarks/vscode:/vscode.git/clone" did not exist on "fe718fd29545dfdaf971c73ffafe3ccb06a25899"

02 Apr, 2026 19 commits
- feat: trace infrastructure — spans + request ID propagation [DIS-1643] (#7733) · 13058cbd
  Neelay Shah authored Apr 02, 2026
```
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
```
  13058cbd
- refactor: move omni post-processing to dedicated formatters / processors (#7746) · 658b0d5a
  Ayush Agarwal authored Apr 02, 2026
```
Signed-off-by: ayushag <ayushag@nvidia.com>
```
  658b0d5a
- Rename stale max_vram_gib marker to profiled_vram_gib (#7829) · 12b4dec5
  Keiven C authored Apr 02, 2026
```
Signed-off-by: Keiven Chang <keivenchang@users.noreply.github.com>
```
  12b4dec5
- feat: support chat_template.json as a prompt formatter artifact (#7785) · 06a24503
  Neal Vaidya authored Apr 02, 2026
```
Closes https://github.com/ai-dynamo/dynamo/issues/7737
```
  06a24503
- fix: support OpenAI content arrays in DeepSeek V3.2 prompt rendering (#6321) · 52a3ca94
  Vladislav Nosivskoy authored Apr 02, 2026
```
Signed-off-by: Vladislav Nosivskoy <vladnosiv@gmail.com>
```
  52a3ca94
- fix: Replace hardcoded SA suffix matching with config-driven allow-list for... · a96d9c51
  Julien Mancuso authored Apr 02, 2026
```
fix: Replace hardcoded SA suffix matching with config-driven allow-list for DGD replicas webhook (#7656) (#7682)
Signed-off-by: Julien Mancuso <jmancuso@nvidia.com>
```
  a96d9c51
- feat: GPU-parallel test runner with VRAM-aware scheduling (#7560) · 6dc85fbc
  Keiven C authored Apr 02, 2026
```
Signed-off-by: Keiven Chang <keivenchang@users.noreply.github.com>
```
  6dc85fbc
- feat(replay): add agg/disagg offline replay optimization [DYN-2566] (#7774) · 4ea21079
  Yan Ru Pei authored Apr 02, 2026
```
Signed-off-by: PeaBrane <yanrpei@gmail.com>
```
  4ea21079
- fix(kv-router): compressed tree sizes accounting (#7800) · b55277c9
  Yan Ru Pei authored Apr 02, 2026
```
Signed-off-by: PeaBrane <yanrpei@gmail.com>
```
  b55277c9
- fix(kv-router): lazily register workers in slot tracker (#7795) · 1bbdff2d
  ishandhanani authored Apr 02, 2026
  
  1bbdff2d
- fix: tool_choice=required bypasses format-specific parsers (#6821) (#7589) · 07af8dca
  Asad Shahid authored Apr 02, 2026
  
  07af8dca
- feat: add sglang embeding cache (#7674) · f46498f2
  Wang, Yi authored Apr 02, 2026
```
Signed-off-by: Wang, Yi <yi.a.wang@intel.com>
```
  f46498f2
- build(snapshot): default CRIU build to upstream criu-dev (#7744) · 93530057
  Schwinn Saereesitthipitak authored Apr 01, 2026
  
  93530057
- feat(vllm): add self-benchmark mode to InstrumentedScheduler (#7779) · 24523a1c
  Hongkuan Zhou authored Apr 01, 2026
  
  24523a1c
- fix: use dict sampling params in prefill warmup (#7797) · a873045c
  weireweire authored Apr 02, 2026
```
Co-authored-by: Weiliangl User <weiliangl@login-node.hosted.internal>
```
  a873045c
- fix: simplify GMS layout state and harden GPU-backed flows (#7006) · 91a8c6af
  Schwinn Saereesitthipitak authored Apr 01, 2026
```
Signed-off-by: mohammedabdulwahhab <furkhan324@berkeley.edu>
Signed-off-by: Hannah Zhang <hannahz@nvidia.com>
Co-authored-by: mohammedabdulwahhab <furkhan324@berkeley.edu>
Co-authored-by: hhzhang16 <54051230+hhzhang16@users.noreply.github.com>
```
  91a8c6af
- feat: add examples and unit tests for vLLM aggregated serving on XPU (#7146) · dd7ceb4a
  Yi Yao authored Apr 02, 2026
```
Signed-off-by: Yi Yao <yi.a.yao@intel.com>
Co-authored-by: Ryan McCormick <rmccormick@nvidia.com>
```
  dd7ceb4a
- refactor(protocols): deprecate cache control (#7790) · c09ac697
  ishandhanani authored Apr 01, 2026
  
  c09ac697
- fix(llm): retry worker-local ZMQ subscriber startup (#7788) · abd0ba5d
  Yan Ru Pei authored Apr 01, 2026
```
Signed-off-by: PeaBrane <yanrpei@gmail.com>
```
  abd0ba5d
01 Apr, 2026 14 commits
- fix: Improve failure handling for profiling failures (#7728) · 546d6a1e
  Ashna Mehrotra authored Apr 01, 2026
```
Signed-off-by: ashnamehrotra <ashnamehrotra@gmail.com>
```
  546d6a1e
- fix: remove forcing the kv block size as 16 in vLLM backend (#7690) · f79b3d62
  Richard Huo authored Apr 01, 2026
  
  f79b3d62
- test: improve deploy test logging for CI debuggability (#7243) · be5b8a58
  Tushar Sharma authored Apr 01, 2026
```
Signed-off-by: Tushar Sharma <tusharma@nvidia.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
```
  be5b8a58
- refactor(2/3): rename dynamo-async-openai to dynamo-protocols (#7565) · b6a3b0c6
  ishandhanani authored Apr 01, 2026
  
  b6a3b0c6
- test: Add ci coverage for trtllm multimodal raw embeddings (#7540) · c84c0934
  Indrajit Bhosale authored Apr 01, 2026
```
Signed-off-by: Indrajit Bhosale <iamindrajitb@gmail.com>
```
  c84c0934
- feat: unified internal request representation for lossless API conversion (#7202) · 3bfee568
  MatejKosec authored Apr 01, 2026
```
Signed-off-by: Matej Kosec <mkosec@nvidia.com>
Signed-off-by: Marko Kosec <mkosec@nvidia.com>
```
  3bfee568
- fix(sglang): use incremental_streaming_output instead of deprecated stream_output (#7642) · 8fe2082c
  YAMY authored Apr 01, 2026
```
Signed-off-by: Yangmin Li <yangminl@nvidia.com>
```
  8fe2082c
- test(planner): isolate planner-family suites [DYN-2534] (#7723) · ab5a31b5
  Alec authored Mar 31, 2026
  
  ab5a31b5
- feat: add least-loaded router (#6314) · cc22114d
  kornelcsernai-harmonic authored Mar 31, 2026
```
Signed-off-by: Kornel Csernai <239206175+kornelcsernai-harmonic@users.noreply.github.com>
Signed-off-by: PeaBrane <yanrpei@gmail.com>
Co-authored-by: PeaBrane <yanrpei@gmail.com>
```
  cc22114d
- fix(backend): Reject conflicting model registrations on the same endpoint (#7686) · 21ba0c4b
  Graham King authored Mar 31, 2026
```
Signed-off-by: Graham King <grahamk@nvidia.com>
```
  21ba0c4b
- feat: propagate OTEL trace context across SGLang multimodal E/P/D workers (#7580) · 8886d184
  jh-nv authored Mar 31, 2026
  
  8886d184
- fix: Flaky cuda event test (#7747) · e5e65b58
  jthomson04 authored Mar 31, 2026
```
Signed-off-by: jthomson04 <jwillthomson19@gmail.com>
```
  e5e65b58
- fix: multimodal ImageLoader errors (#7703) · a8894467
  Qi Wang authored Mar 31, 2026
```
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
```
  a8894467
- refactor: remove dedicated multimodal PD worker (#7556) · 4b1d442b
  GuanLuo authored Mar 31, 2026
```
Signed-off-by: Guan Luo <41310872+GuanLuo@users.noreply.github.com>
```
  4b1d442b
31 Mar, 2026 7 commits
- fix: Support vCluster in post merge in GAIE test (#7743) · b00514c2
  atchernych authored Mar 31, 2026
```
Signed-off-by: Anna Tchernych <atchernych@nvidia.com>
```
  b00514c2
- feat: add initial audio/TTS pipeline support for vLLM-omni backend (#7495) · cbbde3d0
  Zhongdongming Dai authored Mar 31, 2026
```
Signed-off-by: Zhongdongming Dai <zhongdongmin@nvidia.com>
```
  cbbde3d0
- fix: allow having no rust tokenizer when using dyn-chat-processor vllm (#7697) · 76c70f41
  Neal Vaidya authored Mar 31, 2026
  
  76c70f41
- test: split replay tests into a package (#7731) · 75c16393
  Yan Ru Pei authored Mar 31, 2026
```
Signed-off-by: PeaBrane <yanrpei@gmail.com>
```
  75c16393
- feat(frontend): three-layer frontend perf sweep with local and k8s support (#7700) · 273252e6
  Biswa Panda authored Mar 31, 2026
  
  273252e6
- fix: strip null bytes before tokenization to prevent false positive prompt... · 023a299c
  Albert Cheng authored Mar 31, 2026
```
fix: strip null bytes before tokenization to prevent false positive prompt rejection due to Rust/Python tokenizer divergence (#7694)
Signed-off-by: Albert Cheng (Engrg-Hardware 1) <albecheng@login-lyris01.lyris.clusters.nvidia.com>
Co-authored-by: Albert Cheng (Engrg-Hardware 1) <albecheng@login-lyris01.lyris.clusters.nvidia.com>
```
  023a299c
- feat(container): add standalone dynamo-planner image [DYN-2533] (#7696) · a20b9dde
  Alec authored Mar 31, 2026
  
  a20b9dde