 AlpinDale
|
c18bf116da
fix stop strings not being excluded from outputs
|
пре 9 месеци |
|
AlpinDale
|
fe17712f29
fully working chunked prefill
|
пре 9 месеци |
 AlpinDale
|
082d4e6972
feat: add chunked prefill scheduler (#406)
|
пре 9 месеци |
|
AlpinDale
|
f845a661dd
Chunked Prefill Part 2: data update
|
пре 9 месеци |
|
AlpinDale
|
3f5ce50c19
add stop_reason
|
пре 9 месеци |
|
AlpinDale
|
7b9c08afae
vision model support
|
пре 9 месеци |
|
AlpinDale
|
777b6f6d51
add logprob ranks
|
пре 9 месеци |
|
AlpinDale
|
0f1399c135
feat: attention refactor part 2
|
пре 9 месеци |
|
AlpinDale
|
9181fa0396
feat: Triton kernels for sampling (#383)
|
пре 9 месеци |
|
AlpinDale
|
f8dfac6372
chore: attention refactor and upstream sync apr01 (#365)
|
пре 9 месеци |
|
AlpinDale
|
e42a78381a
feat: switch from pylint to ruff (#322)
|
пре 10 месеци |
|
AlpinDale
|
9810daa699
feat: INT8 KV Cache (#298)
|
пре 10 месеци |
|
AlpinDale
|
ac82b67f75
feat: naive context shift and various QoL changes (#289)
|
пре 10 месеци |
|
AlpinDale
|
657aec0cbd
refactor: OpenAI endpoint (#261)
|
пре 10 месеци |
|
AlpinDale
|
4d04ade9ef
feat: fine-grained seeds (#279)
|
пре 11 месеци |
|
AlpinDale
|
d2db4143fa
feat: add grafana for metrics (#240)
|
пре 11 месеци |
|
AlpinDale
|
c0aac15421
feat: S-LoRA support (#222)
|
пре 1 година |
|
AlpinDale
|
8fa608aeb7
feat: replace Ray with NCCL for control plane comms (#221)
|
пре 1 година |
|
AlpinDale
|
f013d714c0
chore: merge dev branch into main (#177)
|
пре 1 година |
|
AlpinDale
|
2755a48d51
merge dev branch into main (#153)
|
пре 1 година |
 50h100a
|
fa0ae5a2c9
feat: new mirostatv2 implementation (#96)
|
пре 1 година |
|
AlpinDale
|
efc6f7fbec
chore: reformats (#90)
|
пре 1 година |
|
AlpinDale
|
e6be0118c9
feat: prompt logprobs and batched samplers (#77)
|
пре 1 година |
|
AlpinDale
|
75c27d3e65
massive overhaul
|
пре 1 година |
|
AlpinDale
|
6dfca14e1f
compute logprobs with log_softmax instead of log
|
пре 1 година |
|
AlpinDale
|
6b9561ef07
adapt TGI incremental detokenization
|
пре 1 година |
|
AlpinDale
|
45f6d9f923
initial refactor commit
|
пре 1 година |
|
AlpinDale
|
f4bb602b74
chore: remove redundant import and minor refactor
|
пре 1 година |
|
AlpinDale
|
c761d38c69
fix: sort outputs and avoid unwanted list copy
|
пре 1 година |
|
AlpinDale
|
7a27bd5f2f
fix: do not allow prompt to exceed max input len
|
пре 1 година |
