Commits · dcbccbcd2ea52d5a0762eb5834718af00317c8e6 · OpenDAS / dynamo · GitLab

12 Mar, 2026 3 commits
- feat(kv-router): add worker_type to router selection log (#7258) · dcbccbcd
  Yan Ru Pei authored Mar 12, 2026
```
Signed-off-by: PeaBrane <yanrpei@gmail.com>
```
  dcbccbcd
- feat(kv-router): package dynamo-kv-indexer binary via maturin [LLM-126] (#7194) · f80ad692
  Yan Ru Pei authored Mar 12, 2026
```
Signed-off-by: PeaBrane <yanrpei@gmail.com>
```
  f80ad692
- feat(kv-router): ZMQ gap detection + replay for standalone indexer [LLM-126] (#7209) · bb07b2f4
  Yan Ru Pei authored Mar 12, 2026
```
Signed-off-by: PeaBrane <yanrpei@gmail.com>
```
  bb07b2f4
11 Mar, 2026 7 commits
- test: pytests able to run locally now (#7219) · 4bd6299b
  Graham King authored Mar 11, 2026
```
Signed-off-by: Graham King <grahamk@nvidia.com>
```
  4bd6299b
- feat: GlobalPlanner --max-total-gpus for cluster-wide GPU budget (#7103) · 5d5fd243
  daiyaanarfeen authored Mar 11, 2026
```
Signed-off-by: Daiyaan <darfeen@nvidia.com>
Signed-off-by: athreesh <anish.maddipoti@utexas.edu>
Signed-off-by: Anish <80174047+athreesh@users.noreply.github.com>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-authored-by: athreesh <anish.maddipoti@utexas.edu>
Co-authored-by: Anish <80174047+athreesh@users.noreply.github.com>
```
  5d5fd243
- test(frontend): Minimal integration test for vllm processor (#7173) · fda022b1
  Graham King authored Mar 11, 2026
```
Signed-off-by: Graham King <grahamk@nvidia.com>
```
  fda022b1
- chore(test): declutter router common.py into helper and router_process modules (#7210) · 96a55928
  Yan Ru Pei authored Mar 11, 2026
```
Signed-off-by: PeaBrane <yanrpei@gmail.com>
```
  96a55928
- fix: serialize disagg first_gen_log_probs int keys for Rust transport (#7145) · e930526b
  Yuewei Na authored Mar 10, 2026
```
Signed-off-by: Yuewei Na <nv-yna@users.noreply.github.com>
Co-authored-by: Yuewei Na <nv-yna@users.noreply.github.com>
```
  e930526b
- fix: Fix mocker+router segfaults + reliability fixes (#7162) · 761e848b
  jthomson04 authored Mar 10, 2026
```
Signed-off-by: jthomson04 <jwillthomson19@gmail.com>
```
  761e848b
- chore: skip lmcache tests on cuda13 env (#7196) · 9eec94b2
  Dmitry Tokarev authored Mar 10, 2026
```
Signed-off-by: Dmitry Tokarev <dtokarev@nvidia.com>
```
  9eec94b2
10 Mar, 2026 10 commits
- feat: split monolithic requirements.txt and remove test deps from runtime image (#6656) · 3cebc864
  Anant Sharma authored Mar 10, 2026
```
Signed-off-by: Anant Sharma <anants@nvidia.com>
```
  3cebc864
- fix: only add profiling data to mocker workers (#7164) · 14d928cb
  Hongkuan Zhou authored Mar 10, 2026
```
Signed-off-by: hongkuanz <hongkuanz@nvidia.com>
```
  14d928cb
- fix: add kv transfer config to planner vllm examples (#7163) · e4865b69
  Hongkuan Zhou authored Mar 10, 2026
```
Signed-off-by: hongkuanz <hongkuanz@nvidia.com>
```
  e4865b69
- fix(perf): Skip duplicate image downloads and unnecessary image processing in... · 6634f33f
  zhongdaor-nv authored Mar 10, 2026
```
fix(perf): Skip duplicate image downloads and unnecessary image processing in MM Router (vLLM) (#7080)
Signed-off-by: zhongdaor <zhongdaor@nvidia.com>
Signed-off-by: zhongdaor-nv <zhongdaor@nvidia.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
```
  6634f33f
- ci: OPS-3802: Increase trtllm timeout, skip unsupported tests (#7104) · 06e19690
  Dillon Cullinan authored Mar 10, 2026
```
Signed-off-by: Dillon Cullinan <dcullinan@nvidia.com>
Signed-off-by: Harrison King Saturley-Hall <hsaturleyhal@nvidia.com>
Co-authored-by: Harrison Saturley-Hall <hsaturleyhal@nvidia.com>
```
  06e19690
- test: update unsupported model name after AIC increase coverage (#7161) · aeaf45f5
  Hongkuan Zhou authored Mar 10, 2026
```
Signed-off-by: hongkuanz <hongkuanz@nvidia.com>
```
  aeaf45f5
- fix: Dont ignore non-zero exitcodes after running tests (#7149) · 09c09909
  Dillon Cullinan authored Mar 10, 2026
```
Signed-off-by: Dillon Cullinan <dcullinan@nvidia.com>
Signed-off-by: Anant Sharma <anants@nvidia.com>
Co-authored-by: Anant Sharma <anants@nvidia.com>
```
  09c09909
- fix(mocker): align vLLM scheduler with v1 — drop watermark, LIFO preemption, retry loop (#7020) · 236cb17d
  Yan Ru Pei authored Mar 10, 2026
```
Signed-off-by: PeaBrane <yanrpei@gmail.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
```
  236cb17d
- fix(ft-tests): avoid shlex.split corrupting JSON args in ServiceSpec setters (#7012) · 6945d47e
  Tzu-Ling Kan authored Mar 09, 2026
```
Signed-off-by: Tzu-Ling <tzulingk@nvidia.com>
```
  6945d47e
- fix: propagate resolved backend and skip interpolation for aggregated configs (#7106) · 50818575
  hhzhang16 authored Mar 09, 2026
```
Signed-off-by: Hannah Zhang <hannahz@nvidia.com>
```
  50818575
09 Mar, 2026 10 commits
- feat(kv-router): P2P recovery for standalone KV indexer [DYN-2367] (#6934) · 48eb52e7
  Yan Ru Pei authored Mar 09, 2026
```
Signed-off-by: PeaBrane <yanrpei@gmail.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
```
  48eb52e7
- fix: increase process group termination timeout from 2s to 8s (#7132) · df67ce16
  Yuewei Na authored Mar 09, 2026
```
Signed-off-by: Yuewei Na <yna@nvidia.com>
Signed-off-by: Yuewei Na <nv-yna@users.noreply.github.com>
Co-authored-by: Yuewei Na <nv-yna@users.noreply.github.com>
```
  df67ce16
- fix: Fix Intermittent KV router + mocker errors (#7108) · a274ef82
  jthomson04 authored Mar 09, 2026
```
Signed-off-by: jthomson04 <jwillthomson19@gmail.com>
```
  a274ef82
- fix: snapshot children before process group kill to prevent GPU memory leaks (#7122) · fddbb84d
  Yuewei Na authored Mar 09, 2026
```
Signed-off-by: Yuewei Na <yna@nvidia.com>
Signed-off-by: Yuewei Na <nv-yna@users.noreply.github.com>
Co-authored-by: Yuewei Na <nv-yna@users.noreply.github.com>
```
  fddbb84d
- fix: use profiler DGD gen route in naive fallback mode (#7099) · 41b357d8
  Hongkuan Zhou authored Mar 09, 2026
```
Signed-off-by: hongkuanz <hongkuanz@nvidia.com>
```
  41b357d8
- fix: correct planner entrypoint in profiler's planner config gen (#7095) · 8a0657cb
  Hongkuan Zhou authored Mar 09, 2026
```
Signed-off-by: hongkuanz <hongkuanz@nvidia.com>
```
  8a0657cb
- test: skip flaky router mocker tests (#7087) (#7102) · e86e8b12
  Dmitry Tokarev authored Mar 09, 2026
```
Signed-off-by: Anant Sharma <anants@nvidia.com>
Signed-off-by: Dmitry Tokarev <dtokarev@nvidia.com>
Co-authored-by: Anant Sharma <anants@nvidia.com>
```
  e86e8b12
- fix: normalize GPUSKU to AIC system identifier (#6984) · 71f9e7a9
  hhzhang16 authored Mar 09, 2026
```
Signed-off-by: Hannah Zhang <hannahz@nvidia.com>
```
  71f9e7a9
- Revert "test: skip flaky router mocker tests (#7087)" (#7092) · b6a115ba
  Dmitry Tokarev authored Mar 09, 2026
  
  b6a115ba
- test: skip flaky router mocker tests (#7087) · 48b302d3
  Anant Sharma authored Mar 09, 2026
```
Signed-off-by: Anant Sharma <anants@nvidia.com>
Signed-off-by: Dmitry Tokarev <dtokarev@nvidia.com>
Co-authored-by: Dmitry Tokarev <dtokarev@nvidia.com>
```
  48b302d3
07 Mar, 2026 1 commit
- chore: bump trtllm to 1.3.0rc5.post1 (#7040) · 25393fb3
  Harrison Saturley-Hall authored Mar 06, 2026
```
Signed-off-by: Harrison King Saturley-Hall <hsaturleyhal@nvidia.com>
```
  25393fb3
06 Mar, 2026 5 commits
- fix: planner e2e test config file using new CLI format (#7014) · e77646ae
  Hongkuan Zhou authored Mar 06, 2026
```
Signed-off-by: hongkuanz <hongkuanz@nvidia.com>
```
  e77646ae
- chore(frontend): Remove the debug_perf flag used for perf work (#7024) · 76c96c5d
  Graham King authored Mar 06, 2026
```
Signed-off-by: Graham King <grahamk@nvidia.com>
```
  76c96c5d
- fix: ETCD cluster unhealthy when adding new member during E2E test & fix vLLM... · 0f01e724
  Jacky authored Mar 06, 2026
```
fix: ETCD cluster unhealthy when adding new member during E2E test & fix vLLM disagg launch config changed (#6976)
Signed-off-by: Jacky <18255193+kthui@users.noreply.github.com>
```
  0f01e724
- test: fix kvbm disagg determinism test on GB200 by setting the kv block size to 32 (#6980) · e02321f8
  Richard Huo authored Mar 05, 2026
  
  e02321f8
- fix: restore --enforce-disagg to reject requests before prefill router activates (#6957) · f934744f
  Yan Ru Pei authored Mar 05, 2026
```
Signed-off-by: PeaBrane <yanrpei@gmail.com>
```
  f934744f
05 Mar, 2026 1 commit
- fix: revert accidental change of trtllm/multimodal_processor (#6921) · 9c6f36e9
  Qi Wang authored Mar 05, 2026
```
Signed-off-by: Qi Wang <qiwa@nvidia.com>
```
  9c6f36e9
04 Mar, 2026 3 commits
- fix: support TRT-LLM 1.3 apply_mm_hashes API (#6810) · d993f9d3
  zhongdaor-nv authored Mar 04, 2026
```
Signed-off-by: zhongdaor <zhongdaor@nvidia.com>
```
  d993f9d3
- docs: testing and container docs update (#6806) · c55f38dc
  Keiven C authored Mar 04, 2026
```
Signed-off-by: Keiven Chang <keivenchang@users.noreply.github.com>
```
  c55f38dc
- fix: exclude nixl .so files from ai-dynamo-runtime wheel (#6430) · 550464b6
  Anant Sharma authored Mar 04, 2026
```
Signed-off-by: Anant Sharma <anants@nvidia.com>
```
  550464b6