Commits · 3c4adde596c6ad51906a0eb2d2f1fed588a317b2 · OpenDAS / dynamo

29 Aug, 2025 3 commits
- chore: add more reasoning and tool call parsers explicitly, remove unecessary files (#2785) · 295964e3
  nachiketb-nvidia authored Aug 29, 2025
```
Signed-off-by: nachiketb <nachiketb@nvidia.com>
```
  295964e3
- feat: Add Encode Worker and NIXL support to trtllm multimodal flow (#2452) · 781fa062
  Indrajit Bhosale authored Aug 29, 2025
```
Signed-off-by: Indrajit Bhosale <iamindrajitb@gmail.com>
```
  781fa062
- fix: [trtllm] add wait_for_instance before register_llm (#2683) · 8c2072cf
  Alec authored Aug 28, 2025
```
Signed-off-by: alec-flowers <aflowers@nvidia.com>
```
  8c2072cf
28 Aug, 2025 1 commit

feat: add reference setup for dynamo logging in k8s with loki (#2699) · 700d345d

mohammedabdulwahhab authored Aug 27, 2025


Signed-off-by: mohammedabdulwahhab <furkhan324@berkeley.edu>
Co-authored-by: coderabbitai[bot] <136622811+coderabbitai[bot]@users.noreply.github.com>

700d345d

27 Aug, 2025 2 commits
- feat: Sglang metrics labels. (#2679) · bad5d122
  Tzu-Ling Kan authored Aug 27, 2025
  
  bad5d122
- feat: Trtllm metric_labels. (#2666) · ef535edb
  Tzu-Ling Kan authored Aug 26, 2025
  
  ef535edb
26 Aug, 2025 4 commits
- feat: add and enable reasoning and tool parser flags for trtllm and sglang (#2713) · 8064849b
  nachiketb-nvidia authored Aug 26, 2025
  
  8064849b
- fix: sglang -- queue requests until model registration completes (#2701) · dfda6205
  hhzhang16 authored Aug 26, 2025
```
Signed-off-by: Hannah Zhang <hannahz@nvidia.com>
```
  dfda6205
- fix: fix sglang multinode example (#2716) · 36df35e4
  julienmancuso authored Aug 26, 2025
  
  36df35e4
- docs: Update how containers should be built for SGLang examples (#2707) · 5d177b61
  Yunzhou Liu authored Aug 26, 2025
  
  5d177b61
25 Aug, 2025 2 commits
- docs: Simplify sphinx build and table of contents on webpage (#2519) · 766d3f2c
  Ryan McCormick authored Aug 25, 2025
  
  766d3f2c
- feat: enable --dyn-reasoning-parser flag to set reasoning parser for vllm deployments (#2700) · f5a41004
  nachiketb-nvidia authored Aug 25, 2025
  
  f5a41004
22 Aug, 2025 4 commits
- fix: [TRTLLM+ LLAMA4 + Eagle 3] Remove the ‘two-models config’ and set the... · 0bd4995d
  Richard Huo authored Aug 22, 2025
```
fix: [TRTLLM+ LLAMA4 + Eagle 3] Remove the ‘two-models config’ and set the ‘one-model’ solution as the default (#2661)
```
  0bd4995d
- feat: [vLLM] implement cli args for tool and reasoning parsers (#2619) · cbe854fc
  Ayush Agarwal authored Aug 22, 2025
  
  cbe854fc
- fix: Update pytest markers for deepep test. (#2561) · f2089d2e
  Kris Hung authored Aug 22, 2025
  
  f2089d2e
- fix: Update tensorrt_llm to 1.0.0rc6 (#2606) · 9ab37d94
  Tanmay Verma authored Aug 21, 2025
  
  9ab37d94
21 Aug, 2025 6 commits
- fix: --kv-event-config now respects command line (#2627) · b5fa0a34
  Alec authored Aug 21, 2025
  
  b5fa0a34
- feat: h100 wideep instructions fix (#2594) · 7d270333
  ishandhanani authored Aug 21, 2025
  
  7d270333
- fix: `skip-tokenizer-init` by default in sglang (#2595) · 08a3763a
  ishandhanani authored Aug 21, 2025
  
  08a3763a
- feat: Add model label for vllm backend metrics (#2474) · 57728909
  Tzu-Ling Kan authored Aug 21, 2025
```
Co-authored-by: Keiven Chang <keivenchang@users.noreply.github.com>
```
  57728909
- feat: use consistent small models across all deploy examples (#2573) · 4eb25632
  Biswa Panda authored Aug 20, 2025
  
  4eb25632
- docs: change sglang hicache example to use hicache-ratio (#2582) · 26b3b609
  Ziqi Fan authored Aug 20, 2025
  
  26b3b609
20 Aug, 2025 5 commits
- feat: allow user to adjust kv_transfer_config (#2517) · bc290e7c
  Alec authored Aug 20, 2025
  
  bc290e7c
- feat(sglang): allow for multi worker spin up in slurm scripts (#2328) · e31c8790
  ishandhanani authored Aug 20, 2025
  
  e31c8790
- fix: resources in wrong location in vllm disag_router example (#2558) · 539c18cd
  julienmancuso authored Aug 20, 2025
  
  539c18cd
- docs: Add health check section to GPT OSS guide (#2556) · 8380f1bd
  Ryan McCormick authored Aug 20, 2025
  
  8380f1bd
- test: Add deepep test for vllm (#2534) · 536280fc
  Kris Hung authored Aug 19, 2025
  
  536280fc
19 Aug, 2025 5 commits
- feat: Use health check and improve instructions for perf sweeps (#2423) · 86a4a58e
  Tanmay Verma authored Aug 19, 2025
  
  86a4a58e
- docs: Add note for LMCache ARM support (#2535) · b9cbee0b
  ZichengMa authored Aug 19, 2025
```
Co-authored-by: Dmitry Tokarev <dtokarev@nvidia.com>
```
  b9cbee0b
- chore: Finish vllm upgrade to 0.10.1 + cleanup (#2528) · 5f57ea5f
  Dmitry Tokarev authored Aug 19, 2025
  
  5f57ea5f
- docs: Consolidate documentation and fix redundant headings (#2518) · 129a2444
  Anish authored Aug 19, 2025
  
  129a2444
- feat(sglang): unify entry point for SGLang backend architecture (#2493) · 80d8aa19
  ishandhanani authored Aug 18, 2025
  
  80d8aa19
18 Aug, 2025 4 commits
- fix: set frontend defaults (#2484) · 28546bad
  mohammedabdulwahhab authored Aug 18, 2025
  
  28546bad
- fix: do not set default for resources (#2471) · 56e99232
  julienmancuso authored Aug 18, 2025
  
  56e99232
- chore: add pre-deployment profiling results on H200 cluster to unblock planner testing (#2495) · dea0b201
  Hongkuan Zhou authored Aug 18, 2025
```
Co-authored-by: Hannah Zhang <hannahz@nvidia.com>
```
  dea0b201
- docs: Bring back some missed release/0.4.0 doc changes, fix broken links, add... · 844f8819
  Ryan McCormick authored Aug 18, 2025
```
docs: Bring back some missed release/0.4.0 doc changes, fix broken links, add lychee link checker github action (#2482)
```
  844f8819
16 Aug, 2025 2 commits
- docs: Add LMCache in vLLM README.md (#2462) · fb10ffb5
  ZichengMa authored Aug 15, 2025
```
Signed-off-by: ZichengMa <zichengma1225@gmail.com>
```
  fb10ffb5
- fix: revert registering runtime config for trtllm backend (deadlock) (#2464) · eaf42eb0
  Yan Ru Pei authored Aug 15, 2025
```
Signed-off-by: Yan Ru Pei <yanrpei@gmail.com>
```
  eaf42eb0
15 Aug, 2025 2 commits
- chore: bug fixes in pre-deployment sweeping and vllm_v1 planner; expose... · 922850ae
  Hongkuan Zhou authored Aug 15, 2025
```
chore: bug fixes in pre-deployment sweeping and vllm_v1 planner; expose num_d/p to k8s metrics (#2454)
```
  922850ae
- chore(sglang): Warn if --skip-tokenizer-init missing (#2463) · 57454300
  Graham King authored Aug 15, 2025
  
  57454300