Commits · dd31a322ae4311f899d8c8950a2bbc8c1a9f47da · OpenDAS / dynamo

09 Mar, 2025 1 commit
- 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
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 2 commits
- chore: regenerate patch (#29) · ea78a424
  Neelay Shah authored Mar 05, 2025
  
  ea78a424
- refactor: rename triton_distributed to dynemo (#22) · 1af7433b
  Neelay Shah authored Mar 05, 2025
```
Co-authored-by: Graham King <grahamk@nvidia.com>
```
  1af7433b
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

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
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