Commits · c82fe888470187a7ad74bde16951a0b9dbc7d123 · OpenDAS / dynamo

20 Feb, 2026 4 commits

feat: add embedding cache to pd worker (#6061) · c82fe888
Qi Wang authored Feb 20, 2026

c82fe888
feat: default kv-events-config to empty (align with vLLM defaults) (#6404) · 7bbacce1
Alec authored Feb 19, 2026
```
Signed-off-by: alec-flowers <aflowers@nvidia.com>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
```
7bbacce1

feat: e2e mm aware kv cache routing support for trtllm backend (#5480) · 23de4e86

zhongdaor-nv authored Feb 19, 2026


Signed-off-by: zhongdaor <zhongdaor@nvidia.com>
Signed-off-by: zhongdaor-nv <zhongdaor@nvidia.com>
Signed-off-by: Zhongdao Ren <zhongdaor@zhongdaor-mlt.client.nvidia.com>
Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
Co-authored-by: Zhongdao Ren <zhongdaor@zhongdaor-mlt.client.nvidia.com>

23de4e86

fix: Fix async pytests - added missing marker (#6439) · 121d8050
Dmitry Tokarev authored Feb 19, 2026
```
Signed-off-by: Dmitry Tokarev <dtokarev@nvidia.com>
```
121d8050

19 Feb, 2026 3 commits
- feat: Add runtime.endpoint() method to eliminate namespace chaining (#6386) · 0ce3461a
  Tzu-Ling Kan authored Feb 19, 2026
```
Signed-off-by: tzulingk@nvidia.com <tzulingk@nvidia.com>
```
  0ce3461a
- test: add vllm audio tests to nightly ci pipeline (#6392) · 6f4b33f7
  knarangN authored Feb 19, 2026
```
Signed-off-by: Kavita Narang <knarang@nvidia.com>
```
  6f4b33f7
- refactor: update frontend kv-router flags to be consistent with router (#6361) · 44a76f96
  jh-nv authored Feb 18, 2026
  
  44a76f96
18 Feb, 2026 6 commits
- feat: planner argparse CLI -> config file (#6356) · b2075619
  Hongkuan Zhou authored Feb 18, 2026
```
Signed-off-by: hongkuanz <hongkuanz@nvidia.com>
```
  b2075619
- test: Add multimodal video tests to nightly CI pipeline (#6023) · 638d8e68
  knarangN authored Feb 18, 2026
```
Signed-off-by: Kavita Narang <knarang@nvidia.com>
```
  638d8e68
- fix: reduce pytest-marker-report output noise and move to tests/ (#6359) · c02cefb5
  Alec authored Feb 18, 2026
```
Signed-off-by: alec-flowers <aflowers@nvidia.com>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
```
  c02cefb5
- feat: add Anthropic Messages API endpoint (/v1/messages) (#6231) · b6603d90
  ishandhanani authored Feb 18, 2026
```
Signed-off-by: Matej Kosec <mkosec@nvidia.com>
Signed-off-by: Marko Kosec <mkosec@nvidia.com>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Co-authored-by: Matej Kosec <mkosec@nvidia.com>
```
  b6603d90
- test: relocate test output to /tmp to keep git working tree clean (#6289) · d86937f9
  Alec authored Feb 17, 2026
```
Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
```
  d86937f9
- refactor: delete handlers and disagg EC producer/consumer (#6051) · d2a57839
  Qi Wang authored Feb 17, 2026
  
  d2a57839
17 Feb, 2026 1 commit

feat: add GlobalPlanner component for centralized scaling (#5702) · 0a266653

daiyaanarfeen authored Feb 17, 2026


Signed-off-by: Daiyaan <darfeen@nvidia.com>
Signed-off-by: daiyaanarfeen <darfeen@nvidia.com>
Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>

0a266653

14 Feb, 2026 3 commits
- feat: load-based scaling in SLA Planner (#6145) · 359765d3
  Hongkuan Zhou authored Feb 14, 2026
```
Signed-off-by: hongkuanz <hongkuanz@nvidia.com>
```
  359765d3
- feat: mm aware routing for vllm (#6235) · 815b1291
  zhongdaor-nv authored Feb 13, 2026
```
Signed-off-by: zhongdaor <zhongdaor@nvidia.com>
Signed-off-by: zhongdaor-nv <zhongdaor@nvidia.com>
```
  815b1291
- feat: migrate GPU discovery from Dynamo Profiler to Dynamo Operator with... · d56439ec
  hhzhang16 authored Feb 13, 2026
```
feat: migrate GPU discovery from Dynamo Profiler to Dynamo Operator with automatic injection (#6224)
Signed-off-by: Hannah Zhang <hannahz@nvidia.com>
```
  d56439ec
13 Feb, 2026 4 commits
- Rename fetch_llm to fetch_model (#6268) · 5624d144
  Tzu-Ling Kan authored Feb 13, 2026
```
Signed-off-by: tzulingk@nvidia.com <tzulingk@nvidia.com>
```
  5624d144
- fix: consolidate dyn_discovery_backend and dyn_kv_store (#6167) · a289695c
  mohammedabdulwahhab authored Feb 13, 2026
```
Signed-off-by: mohammedabdulwahhab <furkhan324@berkeley.edu>
```
  a289695c
- chore: rename KvPushRouter to KvRouter in python + more bindings removal (#6238) · 14eceb43
  Yan Ru Pei authored Feb 12, 2026
```
Signed-off-by: PeaBrane <yanrpei@gmail.com>
```
  14eceb43
- feat: use dynamic port allocation for DYN_SYSTEM_PORT in e2e router t… (#6262) · 166e1f4d
  Keiven C authored Feb 12, 2026
```
Signed-off-by: Keiven Chang <keivenchang@users.noreply.github.com>
```
  166e1f4d
12 Feb, 2026 6 commits

chore: Remove dynamo-run and mistral-rs engine (#6203) · bbe82f18
Graham King authored Feb 12, 2026
```
Signed-off-by: Graham King <grahamk@nvidia.com>
```
bbe82f18
feat: support AIC DGD gen call (WILL BREAK DGDR) (#6216) · a04b5631
Hongkuan Zhou authored Feb 12, 2026
```
Signed-off-by: hongkuanz <hongkuanz@nvidia.com>
```
a04b5631

feat: Flash Indexer (#5785) · 937398cf

Yan Ru Pei authored Feb 11, 2026


Signed-off-by: PeaBrane <yanrpei@gmail.com>
Signed-off-by: jthomson04 <jothomson@nvidia.com>
Signed-off-by: jthomson04 <jwillthomson19@gmail.com>
Signed-off-by: Janelle Cai <jcai18@mit.edu>
Co-authored-by: jthomson04 <jwillthomson19@gmail.com>
Co-authored-by: Janelle Cai <jcai18@mit.edu>

937398cf

feat: responses API compliance with upstream type alignment (#6089) · 8cb47d04
MatejKosec authored Feb 11, 2026
```
Signed-off-by: Matej Kosec <mkosec@nvidia.com>
Co-authored-by: Ishan Dhanani <ishandhanani@gmail.com>
```
8cb47d04
test: change KVBM vLLM interface tests to gpu_0 marker (#6205) · f8d0a9f9
Alec authored Feb 11, 2026
```
Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
```
f8d0a9f9
docs: migrate Fern docs from fern/ into docs/ (#6206) · 39d645e5
Jonathan Tong authored Feb 11, 2026
```
Signed-off-by: Jont828 <jt572@cornell.edu>
```
39d645e5

11 Feb, 2026 7 commits
- fix: use tempfile.TemporaryDirectory in load generator test (#6196) · 45bc1b79
  MatejKosec authored Feb 11, 2026
```
Signed-off-by: Matej Kosec <mkosec@nvidia.com>
```
  45bc1b79
- fix: Cleanup pytest markers, enable gpu_0 tests on trtllm arm, reduce log noise (#6124) · 4220771f
  Dmitry Tokarev authored Feb 11, 2026
```
Signed-off-by: Dmitry Tokarev <dtokarev@nvidia.com>
```
  4220771f
- test: adding interface contract tests for KVBM vLLM integration (#5847) · f9d20c10
  Olga Andreeva authored Feb 11, 2026
  
  f9d20c10
- chore: Templating Feedback Followup (#6125) · 3188c70a
  Dillon Cullinan authored Feb 11, 2026
```
Signed-off-by: Dillon Cullinan <dcullinan@nvidia.com>
```
  3188c70a
- feat: add Prometheus auto and custom label injection for engine metrics (#5989) · e18840ce
  Keiven C authored Feb 10, 2026
```
Signed-off-by: Keiven Chang <keivenchang@users.noreply.github.com>
```
  e18840ce
- test: refactor router e2e tests to use context managers for process lifecycle (#6088) · 6fe2152b
  Karen Chung authored Feb 10, 2026
  
  6fe2152b
- fix: harden KVBM integration tests with dynamic timeouts and metrics (#6105) · 2eb9e0dd
  Keiven C authored Feb 10, 2026
```
Signed-off-by: Keiven Chang <keivenchang@users.noreply.github.com>
```
  2eb9e0dd
10 Feb, 2026 5 commits

test: streamline Python test structure (#5684) · bd344cf9
Qi Wang authored Feb 10, 2026

bd344cf9

fix: prevent aiperf pipe hang in planner scaling test (#6099) · 4ad739dd

MatejKosec authored Feb 10, 2026

Replace PIPE-based stdout/stderr capture with direct file output in LoadGenerator.generate_load() to prevent orphaned aiperf child processes from blocking communicate() indefinitely
Add start_new_session=True so os.killpg() can kill the entire process tree on timeout (not just the main process)
Add unit test validating process-group kill on timeout
Fixes DYN-2086

4ad739dd

feat: Dockerfile templating (#5633) · ac020629
Dillon Cullinan authored Feb 10, 2026
```
Signed-off-by: Dillon Cullinan <dcullinan@nvidia.com>
```
ac020629
fix: increase router E2E timeouts in preparation for enabling xdist (parallel pytest) later (#6085) · 5755a8de
Keiven C authored Feb 09, 2026
```
Signed-off-by: Keiven Chang <keivenchang@users.noreply.github.com>
```
5755a8de
feat: expose Python Prometheus metric via DynamoComponentMetrics (#5817) · 027d2653
Keiven C authored Feb 09, 2026
```
Signed-off-by: Keiven Chang <keivenchang@users.noreply.github.com>
```
027d2653

08 Feb, 2026 1 commit

chore: enable local indexers by default, and use normal event plane by default... · 6783bdca

Yan Ru Pei authored Feb 08, 2026


chore: enable local indexers by default, and use normal event plane by default (not jetstream) (#5941)
Signed-off-by: PeaBrane <yanrpei@gmail.com>

6783bdca