Commits · 9bc6d98d58f10f9671b71c72178fa46affa58c8e · OpenDAS / dynamo · GitLab

13 Apr, 2026 14 commits
- fix(test): unskip test_smoke by switching to trtllm attention backend (DYN-2607) (#8104) · 9bc6d98d
  Yuewei Na authored Apr 13, 2026
```
Signed-off-by: Yuewei Na <nv-yna@users.noreply.github.com>
Co-authored-by: Yuewei Na <nv-yna@users.noreply.github.com>
```
  9bc6d98d
- feat: add KV cache transfer estimated latency metric for disaggregated serving (#7590) · 1ebe7060
  Jonathan Tong authored Apr 13, 2026
```
Signed-off-by: Jont828 <jt572@cornell.edu>
Co-authored-by: Hongkuan Zhou <tedzhouhk@gmail.com>
```
  1ebe7060
- feat: adds support to pass local tokenizer path (#8128) · e41e1715
  yunzhoul-nv authored Apr 13, 2026
```
Signed-off-by: Yunzhou Liu <232973175+yunzhoul-nv@users.noreply.github.com>
```
  e41e1715
- feat(runtime): add DYN_TCP_RESPONSE_STREAM_PORT/HOST env vars for configurable... · 3acdc0f8
  Biswa Panda authored Apr 13, 2026
```
feat(runtime): add DYN_TCP_RESPONSE_STREAM_PORT/HOST env vars for configurable response stream server port
```
  3acdc0f8
- fix: populate input_tokens in Anthropic streaming + add model retrieval endpoint (#7234) · f7fe8005
  MatejKosec authored Apr 13, 2026
```
Signed-off-by: Matej Kosec <mkosec@nvidia.com>
```
  f7fe8005
- docs(router): split router docs into focused pages (#8122) · 49087845
  Yan Ru Pei authored Apr 13, 2026
```
Signed-off-by: PeaBrane <yanrpei@gmail.com>
```
  49087845
- chore: upgrade sglang dependency from 0.5.9 to 0.5.10.post1 (#7997) · e48de6aa
  Krishnan Prashanth authored Apr 13, 2026
```
Signed-off-by: KrishnanPrash <140860868+KrishnanPrash@users.noreply.github.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
```
  e48de6aa
- fix(kvbm): handle typed KvCacheConnectorConfig in consolidator check (#8117) · edce7d3d
  Alec authored Apr 13, 2026
```
Signed-off-by: Alec Flowers <aflowers@nvidia.com>
```
  edce7d3d
- feat: route requests by device type and load for sglang epd (#7215) · a7d51b39
  Wang, Yi authored Apr 14, 2026
```
Signed-off-by: Wang, Yi <yi.a.wang@intel.com>
```
  a7d51b39
- feat(kvbm): build kvbm wheel with nccl feature for CUDA containers (#8120) · a3e6468d
  Alec authored Apr 13, 2026
```
Signed-off-by: Alec Flowers <aflowers@nvidia.com>
```
  a3e6468d
- fix(kv-router): wake queue on remote lifecycle (#8097) · 36e8b8c8
  Yan Ru Pei authored Apr 13, 2026
```
Signed-off-by: PeaBrane <yanrpei@gmail.com>
```
  36e8b8c8
- docs: add GLM-5 NVFP4 recipe to recipes README (#8098) · a0c7025d
  Alec authored Apr 13, 2026
```
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
```
  a0c7025d
- fix(frontend): classify cancelled requests as 499 instead of 500 (#8090) · 1284237e
  Thomas Montfort authored Apr 13, 2026
```
Signed-off-by: tmontfort <tmontfort@nvidia.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
```
  1284237e
- chore(multimodal): Add XPU aggregated video vLLM launch example (#7855) · 934b49ef
  Yi Yao authored Apr 13, 2026
```
Signed-off-by: Yi Yao <yi.a.yao@intel.com>
```
  934b49ef
11 Apr, 2026 5 commits
- feat: Add glm-5 nvfp4 recipe (#7780) · 59df10d1
  William Arnold authored Apr 10, 2026
```
Signed-off-by: William Arnold <warnold@nvidia.com>
```
  59df10d1
- fix: LoRA unregister drops all the LoRA variants that share the same ID (#8053) · 9efa460c
  Keiven C authored Apr 10, 2026
```
Signed-off-by: Keiven Chang <keivenchang@users.noreply.github.com>
Co-authored-by: Keiven Chang <keivenchang@users.noreply.github.com>
```
  9efa460c
- fix(llm): support reading eos_token_ids from tokenizer_config.json for models... · b425b65c
  Ryan McCormick authored Apr 10, 2026
```
fix(llm): support reading eos_token_ids from tokenizer_config.json for models like Qwen3.5 with <|im_end|> token (#8091)
```
  b425b65c
- feat(planner): add diagnostics metrics, rename to dynamo_planner_*, and add HTML report (#8078) · f8920708
  Hongkuan Zhou authored Apr 10, 2026
  
  f8920708
- feat: Request Migration disable on max seq len exceeded (#8020) · f6976e7f
  Jacky authored Apr 10, 2026
```
Signed-off-by: Jacky <18255193+kthui@users.noreply.github.com>
```
  f6976e7f
10 Apr, 2026 17 commits
- fix(ci): multimodal vLLM nightly — predownload, audio KV, 7B toolcalling (DYN-2569) (#7964) · eb772381
  knarangN authored Apr 10, 2026
```
Signed-off-by: Kavita Narang <knarang@nvidia.com>
```
  eb772381
- fix: fix glm5 custom tokenizer loading issue (#8079) · 8ad8f955
  Richard Huo authored Apr 10, 2026
  
  8ad8f955
- fix(operator): prevent orphaned old worker DCDs after rolling update (#7939) · 0488e1b2
  Thomas Montfort authored Apr 10, 2026
```
Signed-off-by: tmontfort <tmontfort@nvidia.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
```
  0488e1b2
- feat(frontend): ResponseTimeout error type with request-plane worker quarantine (#8011) · 7f8e88fb
  MatejKosec authored Apr 10, 2026
```
Signed-off-by: Matej Kosec <mkosec@nvidia.com>
```
  7f8e88fb
- chore: update trtllm version to v1.3.0rc11 (#8048) · a2bc6d5c
  Richard Huo authored Apr 10, 2026
  
  a2bc6d5c
- fix(frontend): backend stream inactivity timeout to fix inflight gauge leak (#7929) · adae833e
  MatejKosec authored Apr 10, 2026
```
Signed-off-by: Matej Kosec <mkosec@nvidia.com>
```
  adae833e
- docs(observability): add cross-engine metrics comparison (#8055) · d1a94558
  Keiven C authored Apr 10, 2026
  
  d1a94558
- feat(fern): add agents page-directive and bump fern version (#8071) · fd8410da
  Ryan Amirthan Stephen authored Apr 10, 2026
```
Signed-off-by: Ryan Amirthan Stephen <ryanstep@umich.edu>
```
  fd8410da
- test: skip persistent nightly CI failures and provide associated tickets (#8006) · bcd0fc06
  Tushar Sharma authored Apr 10, 2026
```
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
```
  bcd0fc06
- docs: Add EFA support steps (#7908) · 1bf3dd5d
  Indrajit Bhosale authored Apr 10, 2026
```
Signed-off-by: Indrajit Bhosale <iamindrajitb@gmail.com>
```
  1bf3dd5d
- feat: generic tokenizer_manager passthrough route for RL training (#6836) · 9412bdff
  William Arnold authored Apr 10, 2026
```
Signed-off-by: William Arnold <warnold@nvidia.com>
```
  9412bdff
- fix: recovery etc order, full recovery, thundering herd on same worker ids in same order. (#8016) · 6b75d6b0
  Michael Feil authored Apr 10, 2026
```
Signed-off-by: michaelfeil <63565275+michaelfeil@users.noreply.github.com>
```
  6b75d6b0
- fix(container): Bump EFA installer version to 1.47.0 (#8061) · 6c877e4d
  Erez Zarum authored Apr 10, 2026
```
Signed-off-by: Erez Zarum <erezz@amazon.com>
```
  6c877e4d
- refactor: inline env var lookups to avoid noisy tracebacks (#8045) · 386aa114
  Dmitry Tokarev authored Apr 10, 2026
```
Signed-off-by: Dmitry Tokarev <dtokarev@nvidia.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
```
  386aa114
- fix: prevent KV block leak from cancel during disagg KV transfer (#7489) · 9055b2d3
  Yuewei Na authored Apr 09, 2026
```
Signed-off-by: Yuewei Na <nv-yna@users.noreply.github.com>
Co-authored-by: Yuewei Na <nv-yna@users.noreply.github.com>
```
  9055b2d3
- fix: add worker dump throtteling (#8013) · 859bd6ea
  Michael Feil authored Apr 09, 2026
```
Signed-off-by: michaelfeil <63565275+michaelfeil@users.noreply.github.com>
```
  859bd6ea
- fix(snapshot): use watch-based waits in snapshotctl (#8024) · a0ecd4dd
  Schwinn Saereesitthipitak authored Apr 09, 2026
```
Signed-off-by: Schwinn Saereesitthipitak <schwinns@nvidia.com>
```
  a0ecd4dd
09 Apr, 2026 4 commits
- fix(profiler): use final TRT-LLM paged KV cache log entry for max tokens (#7988) · e319cd2b
  nihalmaddala authored Apr 09, 2026
```
Signed-off-by: nihalmaddala <nihalmaddala@gmail.com>
```
  e319cd2b
- fix(examples): correct componentType and ServiceAccount for GlobalPlanner in examples (#7755) · 2059f977
  Yao Qing authored Apr 10, 2026
```
Signed-off-by: Yao, Qing <qing.yao@intel.com>
```
  2059f977
- refactor(planner): extract discrete-event state machine with explicit inputs (#8046) · f7e0b3fd
  Hongkuan Zhou authored Apr 09, 2026
  
  f7e0b3fd
- refactor: simplify GPU Memory Service integrations and module boundaries (#7875) · 39a6a240
  Schwinn Saereesitthipitak authored Apr 09, 2026
  
  39a6a240