.. |
attention
|
f6250c5516
move dockerfiles to root; fix cpu build
|
7 kuukautta sitten |
common
|
656459fd84
make fp8_e4m3 work on nvidia
|
7 kuukautta sitten |
distributed
|
5b0c11d190
support pipeline parallel pynccl groups
|
7 kuukautta sitten |
endpoints
|
fe431bb840
check for next port if current is unavailable
|
7 kuukautta sitten |
engine
|
5b0c11d190
support pipeline parallel pynccl groups
|
7 kuukautta sitten |
executor
|
5b0c11d190
support pipeline parallel pynccl groups
|
7 kuukautta sitten |
kv_quant
|
e42a78381a
feat: switch from pylint to ruff (#322)
|
1 vuosi sitten |
lora
|
5b0c11d190
support pipeline parallel pynccl groups
|
7 kuukautta sitten |
modeling
|
5b0c11d190
support pipeline parallel pynccl groups
|
7 kuukautta sitten |
processing
|
5b0c11d190
support pipeline parallel pynccl groups
|
7 kuukautta sitten |
quantization
|
5b0c11d190
support pipeline parallel pynccl groups
|
7 kuukautta sitten |
spec_decode
|
5b0c11d190
support pipeline parallel pynccl groups
|
7 kuukautta sitten |
task_handler
|
5b0c11d190
support pipeline parallel pynccl groups
|
7 kuukautta sitten |
transformers_utils
|
60e74e92fd
add rope_scaling arg
|
7 kuukautta sitten |
__init__.py
|
be8154a8a0
feat: proper embeddings API with e5-mistral-7b support
|
7 kuukautta sitten |
py.typed
|
1c988a48b2
fix logging and add py.typed
|
1 vuosi sitten |