AlpinDale
|
f6250c5516
move dockerfiles to root; fix cpu build
|
6 miesięcy temu |
AlpinDale
|
a94de94c44
refactor: combine the prefill and decode into a single API (#553)
|
6 miesięcy temu |
AlpinDale
|
50b7c13db0
refactor: attention selector (#552)
|
6 miesięcy temu |
AlpinDale
|
0e062e66d3
set block size at init
|
6 miesięcy temu |
AlpinDale
|
35ae01d7ba
refactor: attention metadata term
|
6 miesięcy temu |
AlpinDale
|
aed64884c6
feat: prompt logprobs with chunked prefill (#539)
|
6 miesięcy temu |
AlpinDale
|
fca911ee0a
vLLM Upstream Sync (#526)
|
7 miesięcy temu |
AlpinDale
|
9d81716bfd
[v0.5.3] Release Candidate (#388)
|
8 miesięcy temu |