"vllm/executor/ray_xpu_executor.py" did not exist on "eb6d3c264d0cd8e44dec16bca7947fbe96415ce9"
- 05 Jun, 2025 1 commit
-
-
Tanmay Verma authored
-
- 16 May, 2025 1 commit
-
-
Ryan McCormick authored
-
- 02 May, 2025 1 commit
-
-
Tanmay Verma authored
-
- 28 Apr, 2025 1 commit
-
-
richardhuo-nv authored
We were observing a 40% performance drop compared with trtllm serve when benchmarking with isl=1000 and osl=200 at a concurrency level > 128. The number of the tokenization worker is the bottleneck. After bumping the tokenization processors number to 5, dynamo's benchmarking perf could match the trtllm serve's perf.
-
- 11 Apr, 2025 1 commit
-
-
Tanmay Verma authored
Signed-off-by:
Tanmay Verma <tanmay2592@gmail.com> Signed-off-by:
Tanmay Verma <tanmayv@nvidia.com> Co-authored-by:
Neelay Shah <neelays@nvidia.com>
-
- 10 Apr, 2025 1 commit
-
-
Ziqi Fan authored
-
- 03 Apr, 2025 1 commit
-
-
Tanmay Verma authored
Co-authored-by:
-
- 19 Mar, 2025 1 commit
-
-
ishandhanani authored
-
- 17 Mar, 2025 4 commits
-
-
Neelay Shah authored
-
Anant Sharma authored
-
nnshah1 authored
-
ishandhanani authored
-
- 15 Mar, 2025 2 commits
-
-
Biswa Panda authored
-
Biswa Panda authored
Co-authored-by:
Ryan McCormick <rmccormick@nvidia.com> Co-authored-by:
-
- 14 Mar, 2025 1 commit
-
-
ishandhanani authored
Co-authored-by:Biswa Panda <biswa.panda@gmail.com>
-
