Commits · 761ff073916eeec726a8efcb547f13bfcd2d9a5c · OpenDAS / dynamo

16 Mar, 2025 2 commits
- chore: update version for patched vLLM wheel (#193) · 761ff073
  Anant Sharma authored Mar 16, 2025
  
  761ff073
- chore: adding kubernetes - works for target dev (#192) · 233f782d
  Neelay Shah authored Mar 16, 2025
  
  233f782d
15 Mar, 2025 3 commits
- chore: Apply patch to vLLM wheel (#177) · a2773f3e
  ptarasiewiczNV authored Mar 15, 2025
  
  a2773f3e
- fix: Specify vLLM prebuilt wheel location (#176) · 177fb356
  ptarasiewiczNV authored Mar 15, 2025
  
  177fb356
- fix: fix helm chart deployment (#172) · ffd47bca
  julienmancuso authored Mar 14, 2025
  
  ffd47bca
12 Mar, 2025 1 commit
- test: add basic support for pytest codeblocks (#117) · fcac394a
  Neelay Shah authored Mar 12, 2025
  
  fcac394a
11 Mar, 2025 3 commits
- feat: add new metrics and simple router cost fn (#88) · 3f84cdad
  Alec authored Mar 11, 2025
  
  3f84cdad
- fix: update vLLM patch to 0aa204 (#92) · b4281383
  ptarasiewiczNV authored Mar 11, 2025
  
  b4281383
- chore: Simplify the container build instructions for LLMAPI example (#87) · f784b36a
  Tanmay Verma authored Mar 11, 2025
  
  f784b36a
10 Mar, 2025 1 commit
- build(deps): bump transformers from 4.45.2 to 4.48.3 (#58) · 7591a5cc
  Dmitry Tokarev authored Mar 10, 2025
  
  7591a5cc
09 Mar, 2025 2 commits
- feat: make block_size input for indexer, router, publisher (#66) · 989bb3d5
  Alec authored Mar 09, 2025
  
  989bb3d5
- fix: vLLM disagg fix incorrect block ids order (#63) · 7567620f
  ptarasiewiczNV authored Mar 09, 2025
```
Co-authored-by: ptarasiewicz@nvidia.com <Piotr Tarasiewicz>
```
  7567620f
08 Mar, 2025 1 commit
- chore: rename dynamo (#44) · 602352ce
  Neelay Shah authored Mar 08, 2025
```
Co-authored-by: Biswa Panda <biswa.panda@gmail.com>
```
  602352ce
07 Mar, 2025 2 commits
- feat: onboard dynamo-sdk basic and kv-router examples (#20) · aacc5d76
  Biswa Panda authored Mar 07, 2025
```
Co-authored-by: Neelay Shah <neelays@nvidia.com>
```
  aacc5d76
- chore: remove ucx-py from requirements and fix UCX env variable (#46) · 44bde250
  ptarasiewiczNV authored Mar 07, 2025
```
Co-authored-by: ptarasiewicz@nvidia.com <Piotr Tarasiewicz>
```
  44bde250
06 Mar, 2025 1 commit
- feat: Enable make_xfer NIXL kv transfer (#39) · bc42616e
  ptarasiewiczNV authored Mar 07, 2025
```
Co-authored-by: ptarasiewicz@nvidia.com <Piotr Tarasiewicz>
```
  bc42616e
05 Mar, 2025 4 commits
- chore: regenerate patch (#29) · ea78a424
  Neelay Shah authored Mar 05, 2025
  
  ea78a424
- refactor: updated command line arg (#26) · 3ba2b7e9
  Neelay Shah authored Mar 05, 2025
  
  3ba2b7e9
- refactor: rename triton_distributed to dynemo (#22) · 1af7433b
  Neelay Shah authored Mar 05, 2025
```
Co-authored-by: Graham King <grahamk@nvidia.com>
```
  1af7433b
- feat: OAI compatible endpoints for TRTLLM (#14) · 2791b9ea
  NVShreyas authored Mar 04, 2025
```
Co-authored-by: Tanmay Verma <tanmayv@nvidia.com>
Co-authored-by: Tanmay Verma <tanmayv@login-eos01.eos.clusters.nvidia.com>
Co-authored-by: Tanmay Verma <tanmay2592@gmail.com>
Co-authored-by: Ryan McCormick <rmccormick@nvidia.com>
```
  2791b9ea
04 Mar, 2025 1 commit
- Update NIXL Dockerfile + vLLM patch with variable TP (#9) · fe83f8aa
  ptarasiewiczNV authored Mar 04, 2025
```
Co-authored-by: Piotr Tarasiewicz Nvidia <ptarasiewicznv@Piotrs-MacBook-Pro.local>
```
  fe83f8aa
01 Mar, 2025 1 commit
- Add lazy import to nixl.py · 5f14e467
  Piotr Marcinkiewicz authored Mar 01, 2025
  
  5f14e467
27 Feb, 2025 1 commit

ptarasiewiczNV authored Feb 27, 2025


Co-authored-by: Piotr Tarasiewicz Nvidia <ptarasiewicznv@Piotrs-MacBook-Pro.local>
Co-authored-by: nnshah1 <neelays@nvidia.com>
Co-authored-by: alec-flowers <aflowers@nvidia.com>

f7a60cba

26 Feb, 2025 1 commit
- fix: Add TritonNcclConnector back to vllm patch (#273) · 85535ba0
  Alec authored Feb 25, 2025
  
  85535ba0
25 Feb, 2025 1 commit

feat: enable metrics polling · 861c5098

GuanLuo authored Feb 25, 2025

Signed-off-by: Meenakshi Sharma <163925564+nvda-mesharma@users.noreply.github.com>
Signed-off-by: Neelay Shah <neelays@nvidia.com>
Co-authored-by: Neelay Shah <neelays@nvidia.com>
Co-authored-by: Ryan Olson <ryanolson@users.noreply.github.com>
Co-authored-by: Meenakshi Sharma <163925564+nvda-mesharma@users.noreply.github.com>
Co-authored-by: Biswa Panda <biswapanda@users.noreply.github.com>
Co-authored-by: Ryan McCormick <rmccormick@nvidia.com>

861c5098

21 Feb, 2025 1 commit
- build: Multi-stage VLLM Build (#225) · 6e0ccccb
  Ryan McCormick authored Feb 21, 2025
  
  6e0ccccb
20 Feb, 2025 1 commit

feat: use vllm out of process engine · 60a73634

ptarasiewiczNV authored Feb 20, 2025


Signed-off-by: Piotr Marcinkiewicz <piotrm@nvidia.com>
Co-authored-by: Piotr Marcinkiewicz <piotrm@nvidia.com>
Co-authored-by: Ryan McCormick <rmccormick@nvidia.com>

60a73634

18 Feb, 2025 1 commit

feat: Add KV publisher and receiver. Add KV aware routing example. · 8588e33a

GuanLuo authored Feb 18, 2025


Signed-off-by: Neelay Shah <neelays@nvidia.com>
Co-authored-by: aflowers <aflowers@nvidia.com>
Co-authored-by: Ryan McCormick <rmccormick@nvidia.com>
Co-authored-by: hongkuanz <hongkuanz@nvidia.com>
Co-authored-by: Neelay Shah <neelays@nvidia.com>

8588e33a

17 Feb, 2025 1 commit
- feat: disaggregation support for deepseek arch models (#196) · 9a7b506c
  ptarasiewiczNV authored Feb 17, 2025
  
  9a7b506c
14 Feb, 2025 1 commit

feat: Add event plane · 6d2abdba

Blazej authored Feb 14, 2025


Signed-off-by: Piotr Marcinkiewicz <piotrm@nvidia.com>
Co-authored-by: Piotr Marcinkiewicz <piotrm@nvidia.com>
Co-authored-by: Neelay Shah <neelays@nvidia.com>

6d2abdba

13 Feb, 2025 1 commit
- feat: vLLM v0.7.2 patch supporting XpYd and heterogeneous P/D parallel configs (#167) · f38aa469
  ptarasiewiczNV authored Feb 13, 2025
```
Co-authored-by: Ryan McCormick <rmccormick@nvidia.com>
Co-authored-by: Neelay Shah <neelays@nvidia.com>
```
  f38aa469
12 Feb, 2025 1 commit
- fix: Fix TRTLLM Backend rebuild · c2a9636c
  Tanmay Verma authored Feb 12, 2025
  
  c2a9636c
11 Feb, 2025 1 commit
- chore: Update base from from 24.12 to 25.01 Triton version (#100) · 20b36843
  Tanmay Verma authored Feb 10, 2025
  
  20b36843
08 Feb, 2025 1 commit
- build: Support rebuilding and replacing trtllm backend (#99) · 2cfb1b6d
  Tanmay Verma authored Feb 07, 2025
  
  2cfb1b6d
05 Feb, 2025 1 commit
- ci: Add Copyright Verification Scripts w/ Automation (#110) · c9130f8f
  J Wyman authored Feb 05, 2025
  
  c9130f8f
01 Feb, 2025 1 commit

docs: Add Llama 70B benchmark reproduction and results · 04e50aba

ptarasiewiczNV authored Feb 01, 2025


Signed-off-by: Neelay Shah <neelays@nvidia.com>
Co-authored-by: Neelay Shah <neelays@nvidia.com>
Co-authored-by: Piotr Marcinkiewicz <piotrm@nvidia.com>
Co-authored-by: Ryan McCormick <rmccormick@nvidia.com>

04e50aba

24 Jan, 2025 1 commit
- feat: Add vLLM workers to LLM example (#41) · f6d723f6
  Blazej authored Jan 24, 2025
```
Add example of LLM disaggregated serving
```
  f6d723f6
23 Jan, 2025 1 commit

refactor: openai server skeleton · 607fac29

Neelay Shah authored Jan 22, 2025


Co-authored-by: rmccorm4 <21284872+rmccorm4@users.noreply.github.com>
Co-authored-by: piotrm-nvidia <62554872+piotrm-nvidia@users.noreply.github.com>

607fac29

22 Jan, 2025 1 commit
- build: update isort known first party config (#50) · c3b84790
  Anant Sharma authored Jan 22, 2025
  
  c3b84790
16 Jan, 2025 1 commit
- test: Add sanity test for vllm.distributed patched modules · 7af0d2ef
  Ryan McCormick authored Jan 15, 2025
  
  7af0d2ef