 AlpinDale
|
5be90c3859
Mamba infrastrucuture support (#586)
|
7 месяцев назад |
|
AlpinDale
|
ae04f57ec1
feat: Pipeline Parallel support (#581)
|
7 месяцев назад |
 AlpinDale
|
cdff8e89f9
feat: introduce `DraftModelRunner`
|
7 месяцев назад |
|
AlpinDale
|
405bb74612
Control plane comms refactor (#573)
|
7 месяцев назад |
|
AlpinDale
|
2c321ce1f2
chore: upgrade to rocm 6.1, update docker
|
7 месяцев назад |
|
AlpinDale
|
25feb1d592
chore: add support for pinning lora adapters in the lru cache
|
7 месяцев назад |
|
AlpinDale
|
7194047318
remove vllm-nccl
|
7 месяцев назад |
|
AlpinDale
|
de62ceb18c
refactor: eliminate parallel worker per-step task scheduling overhead
|
8 месяцев назад |
|
AlpinDale
|
ef733aee43
implement ExecuteModelData to reduce executor complexity
|
8 месяцев назад |
|
AlpinDale
|
2e0b115ce1
move func tracing to utils
|
9 месяцев назад |
|
AlpinDale
|
46159b107a
formatting: pt1
|
9 месяцев назад |
|
AlpinDale
|
fca911ee0a
vLLM Upstream Sync (#526)
|
9 месяцев назад |
|
AlpinDale
|
f894f7b176
Revert "reduce dedupe by wrapping in general worker class"
|
10 месяцев назад |
|
AlpinDale
|
9fff6fb507
reduce dedupe by wrapping in general worker class
|
10 месяцев назад |
|
AlpinDale
|
9d81716bfd
[v0.5.3] Release Candidate (#388)
|
10 месяцев назад |
