Commits · 5cbeb5e7439e38a3e165cb4ac9dfa2e0c3c19108 · OpenDAS / dynamo

"vscode:/vscode.git/clone" did not exist on "71321a07cf8a24a69fa3f8a5ebf2319cc937f1cd"

01 Mar, 2025 1 commit
- Add lazy import to nixl.py · 5f14e467
  Piotr Marcinkiewicz authored Mar 01, 2025
  
  5f14e467
28 Feb, 2025 2 commits

ci: initial public ci for PRs (#241) · 37a8ebaf

Harrison Saturley-Hall authored Feb 28, 2025

Signed-off-by: Harrison Saturley-Hall <454891+saturley-hall@users.noreply.github.com>
Signed-off-by: Meenakshi Sharma <163925564+nvda-mesharma@users.noreply.github.com>
Co-authored-by: Meenakshi Sharma <163925564+nvda-mesharma@users.noreply.github.com>

37a8ebaf

Updates to support DS R1 in TRTLLM example (#301) · 12d73a82
NVShreyas authored Feb 28, 2025

12d73a82

27 Feb, 2025 4 commits
- Update wheel build in Dockerfile.vllm_nixl (#295) · 0a393dcb
  ptarasiewiczNV authored Feb 27, 2025
  
  0a393dcb
- feat: vLLM + NIXL example · f7a60cba
  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
- ci: build wheel from root directory (#274) · ea401e3b
  Anant Sharma authored Feb 27, 2025
  
  ea401e3b
- feat: LLM API example integration (#182) · 4b42b232
  Tanmay Verma authored Feb 27, 2025
```
Co-authored-by: NVShreyas <158103197+NVShreyas@users.noreply.github.com>
```
  4b42b232
26 Feb, 2025 3 commits
- Updated genai-perf for hot-fix for vLLM container · cec8248d
  Piotr Marcinkiewicz authored Feb 26, 2025
  
  cec8248d
- Update Dockerfile to include genai-perf hot fix (#276) · 0db2e6c8
  Piotr Marcinkiewicz authored Feb 26, 2025
```
Signed-off-by: Piotr Marcinkiewicz <piotrm@nvidia.com>
```
  0db2e6c8
- fix: Add TritonNcclConnector back to vllm patch (#273) · 85535ba0
  Alec authored Feb 25, 2025
  
  85535ba0
25 Feb, 2025 3 commits

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

refactor: move libs to lib dir · 08fcd7e9

Neelay Shah authored Feb 24, 2025


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

08fcd7e9

refactor: remove python native runtime · 0bfd9a76
Neelay Shah authored Feb 24, 2025

0bfd9a76

24 Feb, 2025 1 commit
- feat: add rust based tokenizer · 4f6f63cd
  Biswa Panda authored Feb 24, 2025
  
  4f6f63cd
21 Feb, 2025 2 commits
- build: Multi-stage VLLM Build (#225) · 6e0ccccb
  Ryan McCormick authored Feb 21, 2025
  
  6e0ccccb
- Add etcd to vLLM dockerfile (#231) · e6be74ba
  Piotr Marcinkiewicz authored Feb 21, 2025
  
  e6be74ba
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 5 commits
- feat: add openai endpoint to the vllm example (#183) · 9e4a548d
  ptarasiewiczNV authored Feb 18, 2025
```
Co-authored-by: Ryan Olson <rolson@nvidia.com>
Co-authored-by: Ryan McCormick <rmccormick@nvidia.com>
```
  9e4a548d
- [fix] add kv routing to vllm dockerfile · 9c9f0086
  aflowers authored Feb 18, 2025
  
  9c9f0086
- fix: Remove duplicate venv setup in Dockerfile (#205) · 94601e77
  Ryan McCormick authored Feb 18, 2025
  
  94601e77
- 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
- feat: Add Dockerfile.vllm (#184) · d8aada0b
  ptarasiewiczNV authored Feb 18, 2025
  
  d8aada0b
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 2 commits
- chore: Add triton_distributed_rs wheel install to container build (#135) · 139a9a83
  Ryan McCormick authored Feb 11, 2025
  
  139a9a83
- chore: Update base from from 24.12 to 25.01 Triton version (#100) · 20b36843
  Tanmay Verma authored Feb 10, 2025
  
  20b36843
08 Feb, 2025 2 commits
- refactor: Modify rust vllm example to use container · d2e7ae02
  Ryan McCormick authored Feb 07, 2025
```
Co-authored-by: Piotr Tarasiewicz <ptarasiewicz@nvidia.com>
Co-authored-by: Neelay Shah <neelays@nvidia.com>
```
  d2e7ae02
- build: Support rebuilding and replacing trtllm backend (#99) · 2cfb1b6d
  Tanmay Verma authored Feb 07, 2025
  
  2cfb1b6d
06 Feb, 2025 1 commit
- fix: Set default 'latest' tag if git tags fail to get parsed (#118) · 7894967e
  Ryan McCormick authored Feb 05, 2025
  
  7894967e
05 Feb, 2025 1 commit
- ci: Add Copyright Verification Scripts w/ Automation (#110) · c9130f8f
  J Wyman authored Feb 05, 2025
  
  c9130f8f
04 Feb, 2025 3 commits
- build: pip based installation of icp and runtime. Also make tritonserver optional · ccfe101e
  Neelay Shah authored Feb 04, 2025
```
Signed-off-by: Neelay Shah <neelays@nvidia.com>
Co-authored-by: Ryan McCormick <rmccormick@nvidia.com>
```
  ccfe101e
- ci: add rust build and pre-merge checks (#91) · 5701753a
  Anant Sharma authored Feb 04, 2025
  
  5701753a
- chore: updating to mv src directory · b34c6f77
  Neelay Shah authored Feb 03, 2025
  
  b34c6f77
03 Feb, 2025 1 commit
- refactor: worker rename to runtime · 4017bd18
  Neelay Shah authored Feb 03, 2025
  
  4017bd18
01 Feb, 2025 2 commits

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

feat(llm): adding initial TRTLLM disaggregation support · deb6c7e8

ishandhanani authored Feb 01, 2025


Co-authored-by: nnshah1 <neelays@nvidia.com>
Co-authored-by: Ryan McCormick <rmccormick@nvidia.com>

deb6c7e8

25 Jan, 2025 1 commit
- chore: Update build version to v0.1.1 and add Teardown section to LLM example README (#66) · 1c1bd7da
  Ryan McCormick authored Jan 24, 2025
  
  1c1bd7da
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