.. |
guided_decoding
|
fde2cda047
chore: update outlines integration from `FSM` to `Guide`
|
7 сар өмнө |
layers
|
9b4c72a801
feat: support channel-wise quant for w8a8 dynamic per token activation quant
|
7 сар өмнө |
model_loader
|
964aa08a70
fix: serializer log
|
7 сар өмнө |
models
|
da6765c084
feat: lora support for commandr models
|
7 сар өмнө |
__init__.py
|
9d81716bfd
[v0.5.3] Release Candidate (#388)
|
10 сар өмнө |
_custom_op.py
|
6a57861fca
feat: initial XPU support via intel_extension_for_pytorch (#571)
|
7 сар өмнө |
pooling_metadata.py
|
be8154a8a0
feat: proper embeddings API with e5-mistral-7b support
|
7 сар өмнө |
sampling_metadata.py
|
b9a5a0ae79
fix: avoid copying prompt/output tokens if penalties arent used
|
7 сар өмнө |
utils.py
|
9d81716bfd
[v0.5.3] Release Candidate (#388)
|
10 сар өмнө |