 AlpinDale
|
5289c14b24
feat: Asymmetric Tensor Parallel (#594)
|
hai 4 meses |
|
AlpinDale
|
5be90c3859
Mamba infrastrucuture support (#586)
|
hai 4 meses |
|
AlpinDale
|
ae04f57ec1
feat: Pipeline Parallel support (#581)
|
hai 4 meses |
|
AlpinDale
|
6a57861fca
feat: initial XPU support via intel_extension_for_pytorch (#571)
|
hai 5 meses |
|
AlpinDale
|
fe21123a1c
feat: TPU support (#570)
|
hai 5 meses |
 AlpinDale
|
f40b809d3b
allow using v2 block manager with sliding window
|
hai 5 meses |
|
AlpinDale
|
50b7c13db0
refactor: attention selector (#552)
|
hai 5 meses |
|
AlpinDale
|
8b56dc4347
dict -> torch.Tensor for blocks_to_swap
|
hai 5 meses |
|
AlpinDale
|
21ce19b3ea
blocks_to_copy dict -> torch.Tensor
|
hai 5 meses |
|
AlpinDale
|
9d81716bfd
[v0.5.3] Release Candidate (#388)
|
hai 8 meses |
|
AlpinDale
|
e3252edd07
fix: remove event and stream, add typing (#382)
|
hai 9 meses |
|
AlpinDale
|
f8dfac6372
chore: attention refactor and upstream sync apr01 (#365)
|
hai 9 meses |
|
AlpinDale
|
c2d77b1822
chore: logging refactor (#302)
|
hai 10 meses |
|
AlpinDale
|
ea0f57b233
feat: allow further support for non-cuda devices (#247)
|
hai 11 meses |
|
AlpinDale
|
31c95011a6
feat: FP8 E5M2 KV Cache (#226)
|
hai 11 meses |
|
AlpinDale
|
8fa608aeb7
feat: replace Ray with NCCL for control plane comms (#221)
|
hai 11 meses |
|
AlpinDale
|
15a0454172
feat: FP8 KV Cache (#185)
|
hai 1 ano |
|
AlpinDale
|
1aab8a7d6f
feat: speedup compilation times by 3x (#130)
|
hai 1 ano |
|
AlpinDale
|
74604eb252
fix: pylint complaints (#91)
|
hai 1 ano |
|
AlpinDale
|
75c27d3e65
massive overhaul
|
hai 1 ano |
|
AlpinDale
|
525edab7cc
fix: logger in cache engine
|
hai 1 ano |
|
AlpinDale
|
b8f4337c5b
chore: various fixes
|
hai 1 ano |
|
AlpinDale
|
a409431c40
feat: draft for cuda kernels
|
hai 1 ano |
