AlpinDale fe01e2ded8 chore: move `device` keys to a constant (#1020) 2 hónapja
..
adapter_commons 2f61644f6e SPMD optimizations (#824) 3 hónapja
assets 411ac4f405 vlm: add support for Qwen2-VL model (#1015) 2 hónapja
attention de341ffb00 fix: ensure multistep lookahead allocation is compatible with cugraph max capture (#1008) 2 hónapja
common a113309876 kernel: add meta functions for ops to prevent graph breaks (#1019) 2 hónapja
compilation 0e5cf7f840 tpu: avoid dynamo guard eval overhead (#949) 2 hónapja
distributed 61103b92d4 tpu: support single and multi-host TPUs on GKE and RayServe (#970) 2 hónapja
endpoints 411ac4f405 vlm: add support for Qwen2-VL model (#1015) 2 hónapja
engine fe01e2ded8 chore: move `device` keys to a constant (#1020) 2 hónapja
executor f2b6dc3872 cpu: add support for W8A8 quantization via compressed-tensor (#1017) 2 hónapja
inputs 908ff753a1 fix: phi_3.5_v loading (#896) 2 hónapja
kv_quant 8a71788372 Add OLMoE (#772) 4 hónapja
lora bf4a4d8516 fix: do not register punica with torch if using older torch (#948) 2 hónapja
modeling a113309876 kernel: add meta functions for ops to prevent graph breaks (#1019) 2 hónapja
multimodal 411ac4f405 vlm: add support for Qwen2-VL model (#1015) 2 hónapja
platforms f2b6dc3872 cpu: add support for W8A8 quantization via compressed-tensor (#1017) 2 hónapja
plugins 22a4cd4595 core: fix spec decode metrics and envs circular import (#889) 2 hónapja
processing f561a54a43 core: fix async postprocessor in case of preemption (#1000) 2 hónapja
prompt_adapter 30d02d0747 chore: remove peft as a requirement (#1006) 2 hónapja
quantization f2b6dc3872 cpu: add support for W8A8 quantization via compressed-tensor (#1017) 2 hónapja
server 22a4cd4595 core: fix spec decode metrics and envs circular import (#889) 2 hónapja
spec_decode 5c3b94de45 spec decode: move ops.advane_step to flash attention backend (#1005) 2 hónapja
transformers_utils 411ac4f405 vlm: add support for Qwen2-VL model (#1015) 2 hónapja
triton_utils f1d0b77c92 [0.6.0] Release Candidate (#481) 6 hónapja
worker a113309876 kernel: add meta functions for ops to prevent graph breaks (#1019) 2 hónapja
__init__.py f1d0b77c92 [0.6.0] Release Candidate (#481) 6 hónapja
_core_ext.py f1ea7711bd core: do not compile ScalarType for torch < 2.4.0 (#938) 2 hónapja
_custom_ops.py a113309876 kernel: add meta functions for ops to prevent graph breaks (#1019) 2 hónapja
_ipex_ops.py f1d0b77c92 [0.6.0] Release Candidate (#481) 6 hónapja
connections.py c6c91edab7 ci: update & overhaul test units (#769) 3 hónapja
constants.py 2f61644f6e SPMD optimizations (#824) 3 hónapja
py.typed 1c988a48b2 fix logging and add py.typed 1 éve
scalar_type.py f1d0b77c92 [0.6.0] Release Candidate (#481) 6 hónapja
version.py cbd51a208a ci: bump to 0.6.5 (#964) 2 hónapja