AlpinDale
|
3d83e64f8e
feat: add metrics for prefix cache hit rate (#829)
|
1 ماه پیش |
AlpinDale
|
2f61644f6e
SPMD optimizations (#824)
|
1 ماه پیش |
AlpinDale
|
0a369f9171
feat: support chunked prefill with LoRA (#823)
|
1 ماه پیش |
AlpinDale
|
577586309d
chore: multi-step args and sequence modifications (#713)
|
4 ماه پیش |
AlpinDale
|
bf88c8567e
feat: mamba model support (#674)
|
4 ماه پیش |
AlpinDale
|
7df7b8ca53
optimization: reduce end-to-end overhead from python obj allocation (#666)
|
4 ماه پیش |
AlpinDale
|
a0e446a17d
feat: initial encoder-decoder support with BART model (#633)
|
4 ماه پیش |
AlpinDale
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
4 ماه پیش |
AlpinDale
|
9d81716bfd
[v0.5.3] Release Candidate (#388)
|
8 ماه پیش |
AlpinDale
|
78d66f16d1
Chunked Prefill Part 1 (#384)
|
9 ماه پیش |
AlpinDale
|
f8dfac6372
chore: attention refactor and upstream sync apr01 (#365)
|
9 ماه پیش |
AlpinDale
|
e42a78381a
feat: switch from pylint to ruff (#322)
|
10 ماه پیش |
AlpinDale
|
c2d77b1822
chore: logging refactor (#302)
|
10 ماه پیش |
AlpinDale
|
ac82b67f75
feat: naive context shift and various QoL changes (#289)
|
10 ماه پیش |
AlpinDale
|
4d04ade9ef
feat: fine-grained seeds (#279)
|
10 ماه پیش |
AlpinDale
|
c0aac15421
feat: S-LoRA support (#222)
|
11 ماه پیش |
AlpinDale
|
8fa608aeb7
feat: replace Ray with NCCL for control plane comms (#221)
|
11 ماه پیش |
AlpinDale
|
b9b295d74e
chore: backlogs 1 (#191)
|
1 سال پیش |
g4rg
|
2aab3da9bd
chore: fix Python 3.8+ compatibility (#170)
|
1 سال پیش |
AlpinDale
|
9ec4e08ade
fix: cpu sync delay fix (#127)
|
1 سال پیش |
AlpinDale
|
13901af940
fix: scheduler hang with long prompts (#126)
|
1 سال پیش |
50h100a
|
fa0ae5a2c9
feat: new mirostatv2 implementation (#96)
|
1 سال پیش |
AlpinDale
|
efc6f7fbec
chore: reformats (#90)
|
1 سال پیش |
AlpinDale
|
3d72f05c7b
feat: flattened 1D tensor -> 2D tensor (#85)
|
1 سال پیش |
AlpinDale
|
04a27c6aeb
fix: revert mirostat v2 (#79)
|
1 سال پیش |
Stefan Gligorijevic
|
5dbd262033
feat: Mirostat v2 (#69)
|
1 سال پیش |
AlpinDale
|
cbeeabeb9a
feat: mistral support (#20)
|
1 سال پیش |
AlpinDale
|
75c27d3e65
massive overhaul
|
1 سال پیش |
AlpinDale
|
45f6d9f923
initial refactor commit
|
1 سال پیش |
AlpinDale
|
826de3ef93
use flash attention with xformers
|
1 سال پیش |