ipex-llm 2.2.0b20250324__py3-none-manylinux2010_x86_64.whl → 2.2.0b20250326__py3-none-manylinux2010_x86_64.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
@@ -192,7 +192,8 @@ def is_linear_module(module):
192
192
  and hasattr(module.quant_method, "quant_config")
193
193
  and module.quant_method.quant_config.get_name() == "gptq"):
194
194
  _USE_VLLM_GPTQ = True
195
- invalidInputError(module.skip_bias_add is not True, "Currently, ipex-vllm does not"
195
+ invalidInputError(module.skip_bias_add is not True or module.bias is None,
196
+ "Currently, ipex-vllm does not"
196
197
  " support linear layers with skip_bias_add argument")
197
198
  if isinstance(module, RowParallelLinear) and tp_size >= 2:
198
199
  mp_group = get_tensor_model_parallel_group()
@@ -1275,6 +1276,8 @@ def _optimize_post(model):
1275
1276
  convert_forward(model,
1276
1277
  module.BertSelfAttention,
1277
1278
  self_attention_forward)
1279
+ if hasattr(module, "BertSdpaSelfAttention"):
1280
+ convert_forward(model, module.BertSdpaSelfAttention, self_attention_forward)
1278
1281
  convert_forward(model,
1279
1282
  module.BertEncoder,
1280
1283
  encoder_forward)
@@ -1863,6 +1866,8 @@ def _optimize_post(model):
1863
1866
  convert_forward(model,
1864
1867
  module.BertSelfAttention,
1865
1868
  self_attention_forward)
1869
+ if hasattr(module, "BertSdpaSelfAttention"):
1870
+ convert_forward(model, module.BertSdpaSelfAttention, self_attention_forward)
1866
1871
  convert_forward(model,
1867
1872
  module.BertEncoder,
1868
1873
  encoder_forward)
@@ -119,7 +119,7 @@ def encoder_forward(
119
119
  output_hidden_states: Optional[bool] = False,
120
120
  return_dict: Optional[bool] = True,
121
121
  ):
122
- if not attention_mask.any():
122
+ if attention_mask and not attention_mask.any():
123
123
  attention_mask = None
124
124
  return BertEncoder.forward(
125
125
  self=self,
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.1
2
2
  Name: ipex-llm
3
- Version: 2.2.0b20250324
3
+ Version: 2.2.0b20250326
4
4
  Summary: Large Language Model Develop Toolkit
5
5
  Home-page: https://github.com/intel-analytics/ipex-llm
6
6
  Author: BigDL Authors
@@ -27,7 +27,7 @@ Requires-Dist: intel-openmp ; (platform_machine == "x86_64" or platform_machine
27
27
  Requires-Dist: torch ==2.1.2+cpu ; (platform_system == "Linux") and extra == 'all'
28
28
  Requires-Dist: torch ==2.1.2 ; (platform_system == "Windows") and extra == 'all'
29
29
  Provides-Extra: cpp
30
- Requires-Dist: bigdl-core-cpp ==2.6.0b20250324 ; extra == 'cpp'
30
+ Requires-Dist: bigdl-core-cpp ==2.6.0b20250326 ; extra == 'cpp'
31
31
  Requires-Dist: setuptools ; extra == 'cpp'
32
32
  Requires-Dist: onednn-devel ==2025.0.1 ; (platform_system == "Windows") and extra == 'cpp'
33
33
  Requires-Dist: onednn ==2025.0.1 ; (platform_system == "Windows") and extra == 'cpp'
@@ -60,7 +60,7 @@ Requires-Dist: transformers ==4.40.0 ; extra == 'npu'
60
60
  Requires-Dist: intel-openmp ; (platform_machine == "x86_64" or platform_machine == "AMD64") and extra == 'npu'
61
61
  Requires-Dist: torch ==2.1.2+cpu ; (platform_system == "Linux") and extra == 'npu'
62
62
  Requires-Dist: torch ==2.1.2 ; (platform_system == "Windows") and extra == 'npu'
63
- Requires-Dist: bigdl-core-npu ==2.6.0b20250324 ; (platform_system == "Windows") and extra == 'npu'
63
+ Requires-Dist: bigdl-core-npu ==2.6.0b20250326 ; (platform_system == "Windows") and extra == 'npu'
64
64
  Provides-Extra: serving
65
65
  Requires-Dist: py-cpuinfo ; extra == 'serving'
66
66
  Requires-Dist: fschat[model_worker,webui] ==0.2.36 ; extra == 'serving'
@@ -80,9 +80,9 @@ Requires-Dist: setuptools <70.0.0 ; extra == 'xpu'
80
80
  Requires-Dist: torch ==2.1.0a0 ; extra == 'xpu'
81
81
  Requires-Dist: torchvision ==0.16.0a0 ; extra == 'xpu'
82
82
  Requires-Dist: intel-extension-for-pytorch ==2.1.10+xpu ; extra == 'xpu'
83
- Requires-Dist: bigdl-core-xe-21 ==2.6.0b20250324 ; extra == 'xpu'
84
- Requires-Dist: bigdl-core-xe-batch-21 ==2.6.0b20250324 ; extra == 'xpu'
85
- Requires-Dist: bigdl-core-xe-addons-21 ==2.6.0b20250324 ; extra == 'xpu'
83
+ Requires-Dist: bigdl-core-xe-21 ==2.6.0b20250326 ; extra == 'xpu'
84
+ Requires-Dist: bigdl-core-xe-batch-21 ==2.6.0b20250326 ; extra == 'xpu'
85
+ Requires-Dist: bigdl-core-xe-addons-21 ==2.6.0b20250326 ; extra == 'xpu'
86
86
  Provides-Extra: xpu-2-1
87
87
  Requires-Dist: py-cpuinfo ; extra == 'xpu-2-1'
88
88
  Requires-Dist: protobuf ; extra == 'xpu-2-1'
@@ -97,9 +97,9 @@ Requires-Dist: setuptools <70.0.0 ; extra == 'xpu-2-1'
97
97
  Requires-Dist: torch ==2.1.0a0 ; extra == 'xpu-2-1'
98
98
  Requires-Dist: torchvision ==0.16.0a0 ; extra == 'xpu-2-1'
99
99
  Requires-Dist: intel-extension-for-pytorch ==2.1.10+xpu ; extra == 'xpu-2-1'
100
- Requires-Dist: bigdl-core-xe-21 ==2.6.0b20250324 ; extra == 'xpu-2-1'
101
- Requires-Dist: bigdl-core-xe-batch-21 ==2.6.0b20250324 ; extra == 'xpu-2-1'
102
- Requires-Dist: bigdl-core-xe-addons-21 ==2.6.0b20250324 ; extra == 'xpu-2-1'
100
+ Requires-Dist: bigdl-core-xe-21 ==2.6.0b20250326 ; extra == 'xpu-2-1'
101
+ Requires-Dist: bigdl-core-xe-batch-21 ==2.6.0b20250326 ; extra == 'xpu-2-1'
102
+ Requires-Dist: bigdl-core-xe-addons-21 ==2.6.0b20250326 ; extra == 'xpu-2-1'
103
103
  Requires-Dist: intel-openmp ; (platform_machine == "x86_64" or platform_machine == "AMD64") and extra == 'xpu-2-1'
104
104
  Requires-Dist: dpcpp-cpp-rt ==2024.0.2 ; (platform_system == "Windows") and extra == 'xpu-2-1'
105
105
  Requires-Dist: mkl-dpcpp ==2024.0.0 ; (platform_system == "Windows") and extra == 'xpu-2-1'
@@ -117,7 +117,7 @@ Requires-Dist: setuptools ; extra == 'xpu-2-6'
117
117
  Requires-Dist: torch ==2.6.0+xpu ; extra == 'xpu-2-6'
118
118
  Requires-Dist: torchvision ==0.21.0+xpu ; extra == 'xpu-2-6'
119
119
  Requires-Dist: torchaudio ==2.6.0+xpu ; extra == 'xpu-2-6'
120
- Requires-Dist: bigdl-core-xe-all ==2.6.0b20250324 ; extra == 'xpu-2-6'
120
+ Requires-Dist: bigdl-core-xe-all ==2.6.0b20250326 ; extra == 'xpu-2-6'
121
121
  Requires-Dist: onednn-devel ==2025.0.1 ; extra == 'xpu-2-6'
122
122
  Requires-Dist: onednn ==2025.0.1 ; extra == 'xpu-2-6'
123
123
  Requires-Dist: dpcpp-cpp-rt ==2025.0.2 ; extra == 'xpu-2-6'
@@ -132,7 +132,7 @@ Requires-Dist: tokenizers ==0.15.2 ; extra == 'xpu-2-6-arl'
132
132
  Requires-Dist: accelerate ==0.23.0 ; extra == 'xpu-2-6-arl'
133
133
  Requires-Dist: tabulate ; extra == 'xpu-2-6-arl'
134
134
  Requires-Dist: setuptools ; extra == 'xpu-2-6-arl'
135
- Requires-Dist: bigdl-core-xe-all ==2.6.0b20250324 ; extra == 'xpu-2-6-arl'
135
+ Requires-Dist: bigdl-core-xe-all ==2.6.0b20250326 ; extra == 'xpu-2-6-arl'
136
136
  Requires-Dist: onednn-devel ==2025.0.1 ; extra == 'xpu-2-6-arl'
137
137
  Requires-Dist: onednn ==2025.0.1 ; extra == 'xpu-2-6-arl'
138
138
  Requires-Dist: dpcpp-cpp-rt ==2025.0.2 ; extra == 'xpu-2-6-arl'
@@ -155,9 +155,9 @@ Requires-Dist: tokenizers ==0.15.2 ; extra == 'xpu-arc'
155
155
  Requires-Dist: accelerate ==0.23.0 ; extra == 'xpu-arc'
156
156
  Requires-Dist: tabulate ; extra == 'xpu-arc'
157
157
  Requires-Dist: setuptools ; extra == 'xpu-arc'
158
- Requires-Dist: bigdl-core-xe-23 ==2.6.0b20250324 ; extra == 'xpu-arc'
159
- Requires-Dist: bigdl-core-xe-batch-23 ==2.6.0b20250324 ; extra == 'xpu-arc'
160
- Requires-Dist: bigdl-core-xe-addons-23 ==2.6.0b20250324 ; extra == 'xpu-arc'
158
+ Requires-Dist: bigdl-core-xe-23 ==2.6.0b20250326 ; extra == 'xpu-arc'
159
+ Requires-Dist: bigdl-core-xe-batch-23 ==2.6.0b20250326 ; extra == 'xpu-arc'
160
+ Requires-Dist: bigdl-core-xe-addons-23 ==2.6.0b20250326 ; extra == 'xpu-arc'
161
161
  Requires-Dist: intel-openmp ; (platform_machine == "x86_64" or platform_machine == "AMD64") and extra == 'xpu-arc'
162
162
  Requires-Dist: torch ==2.3.1+cxx11.abi ; (platform_system == "Linux") and extra == 'xpu-arc'
163
163
  Requires-Dist: torchvision ==0.18.1+cxx11.abi ; (platform_system == "Linux") and extra == 'xpu-arc'
@@ -178,9 +178,9 @@ Requires-Dist: tokenizers ==0.15.2 ; extra == 'xpu-arl'
178
178
  Requires-Dist: accelerate ==0.23.0 ; extra == 'xpu-arl'
179
179
  Requires-Dist: tabulate ; extra == 'xpu-arl'
180
180
  Requires-Dist: setuptools ; extra == 'xpu-arl'
181
- Requires-Dist: bigdl-core-xe-23 ==2.6.0b20250324 ; extra == 'xpu-arl'
182
- Requires-Dist: bigdl-core-xe-batch-23 ==2.6.0b20250324 ; extra == 'xpu-arl'
183
- Requires-Dist: bigdl-core-xe-addons-23 ==2.6.0b20250324 ; extra == 'xpu-arl'
181
+ Requires-Dist: bigdl-core-xe-23 ==2.6.0b20250326 ; extra == 'xpu-arl'
182
+ Requires-Dist: bigdl-core-xe-batch-23 ==2.6.0b20250326 ; extra == 'xpu-arl'
183
+ Requires-Dist: bigdl-core-xe-addons-23 ==2.6.0b20250326 ; extra == 'xpu-arl'
184
184
  Requires-Dist: intel-openmp ; (platform_machine == "x86_64" or platform_machine == "AMD64") and extra == 'xpu-arl'
185
185
  Requires-Dist: torch ==2.3.1+cxx11.abi ; (platform_system == "Linux") and extra == 'xpu-arl'
186
186
  Requires-Dist: torchvision ==0.18.1+cxx11.abi ; (platform_system == "Linux") and extra == 'xpu-arl'
@@ -201,9 +201,9 @@ Requires-Dist: tokenizers ==0.15.2 ; extra == 'xpu-lnl'
201
201
  Requires-Dist: accelerate ==0.23.0 ; extra == 'xpu-lnl'
202
202
  Requires-Dist: tabulate ; extra == 'xpu-lnl'
203
203
  Requires-Dist: setuptools ; extra == 'xpu-lnl'
204
- Requires-Dist: bigdl-core-xe-23 ==2.6.0b20250324 ; extra == 'xpu-lnl'
205
- Requires-Dist: bigdl-core-xe-batch-23 ==2.6.0b20250324 ; extra == 'xpu-lnl'
206
- Requires-Dist: bigdl-core-xe-addons-23 ==2.6.0b20250324 ; extra == 'xpu-lnl'
204
+ Requires-Dist: bigdl-core-xe-23 ==2.6.0b20250326 ; extra == 'xpu-lnl'
205
+ Requires-Dist: bigdl-core-xe-batch-23 ==2.6.0b20250326 ; extra == 'xpu-lnl'
206
+ Requires-Dist: bigdl-core-xe-addons-23 ==2.6.0b20250326 ; extra == 'xpu-lnl'
207
207
  Requires-Dist: intel-openmp ; (platform_machine == "x86_64" or platform_machine == "AMD64") and extra == 'xpu-lnl'
208
208
  Requires-Dist: torch ==2.3.1+cxx11.abi ; (platform_system == "Linux") and extra == 'xpu-lnl'
209
209
  Requires-Dist: torchvision ==0.18.1+cxx11.abi ; (platform_system == "Linux") and extra == 'xpu-lnl'
@@ -94,7 +94,7 @@ ipex_llm/serving/fastchat/tgi_api_protocol.py,sha256=brT3k3-V0NJrU4fRqUwWjC0O3iO
94
94
  ipex_llm/serving/fastchat/tgi_api_server.py,sha256=agNTAEiZPSuj3dEdIdYKwkoY0cXOUDX06DiM9VP2knQ,24418
95
95
  ipex_llm/serving/fastchat/vllm_worker.py,sha256=ZLz2Q9GxJO6r_LOiP6epgCRjBGk-K4EB1SNEWSJp5DA,11091
96
96
  ipex_llm/transformers/__init__.py,sha256=BreA3EY6hkNq0rVixb_sUuTLzMrcWXTt3yRsshCPHHQ,1214
97
- ipex_llm/transformers/convert.py,sha256=6q2iB-NwtIxLg0mlksef4oLElwZr3stMxH42d16gMWM,103417
97
+ ipex_llm/transformers/convert.py,sha256=CfdQz1wDgjmAT4uT_y9ZRbOrFjmsXxFUgPeWEExskk8,103770
98
98
  ipex_llm/transformers/convert_ipex.py,sha256=_nSnUTQy-yfkKaqGdqnBdWztZf3NGmnbZ0TKaDrF4X4,14617
99
99
  ipex_llm/transformers/embedding.py,sha256=bdgk59DvD4ZZyxRzewXOR7g56nThgO6uhIwk8QL7f-s,9299
100
100
  ipex_llm/transformers/kv.py,sha256=k4TU18LlA-Sbq9WNNQnfuzu3RSFBwFhmaV3BcGN5bAo,19191
@@ -147,7 +147,7 @@ ipex_llm/transformers/models/__init__.py,sha256=tp2DcVkKg1-QvdYk7DY7rZvQWCDQ4ZjU
147
147
  ipex_llm/transformers/models/aquila.py,sha256=VZb5Drpo_fTxwcExZ397LygnsNPX2sVbie9_JeFudZI,5252
148
148
  ipex_llm/transformers/models/baichuan.py,sha256=8b43mBRZJEf_xLNoodhA4r9x1anqwC3Wt8awWel-aUo,18306
149
149
  ipex_llm/transformers/models/baichuan_m1.py,sha256=l6BD9jbA8TQ5Q0fcngHNRw73SxqvC126ErENs7Zh-do,9513
150
- ipex_llm/transformers/models/bert.py,sha256=0Mm9jkvkzBxtc_z_GE1TcZoPz-HOg2Z2973ZEWgSwJk,5601
150
+ ipex_llm/transformers/models/bert.py,sha256=U3JVnuY1z0ILlJgMvetOKuCAkH74ufGbg4whZ8GNtmQ,5620
151
151
  ipex_llm/transformers/models/bloom.py,sha256=PxfzyYT-nFn3K5rZhTQjmcEjUUzAhUFzxIN4kzRlCuc,8103
152
152
  ipex_llm/transformers/models/chatglm.py,sha256=UHai1t2AUtGmF765_eHF8LUMVQzp_oCBx8TJB21WrHk,12597
153
153
  ipex_llm/transformers/models/chatglm2.py,sha256=KyAIX7zGVQDQuwwM3QMBNWZbTeMHEzKUIgAryT0voHc,14933
@@ -262,11 +262,11 @@ ipex_llm/vllm/xpu/engine/__init__.py,sha256=pY_CpyuZd72fr6s32ejeKHKFW0K4vUU2rzZj
262
262
  ipex_llm/vllm/xpu/engine/engine.py,sha256=NvCMbp0X8NVrOqbwm4FTvXOptTRLzu9jQsy37ZHnTk8,9493
263
263
  ipex_llm/vllm/xpu/entrypoints/openai/api_server.py,sha256=IjiSze9vzBCAkLu_VwIcJwuO1jyFna7DLrj6aSL7RaQ,35220
264
264
  ipex_llm/vllm/xpu/entrypoints/openai/cli_args.py,sha256=hB398yYtKauASRzevctScdbFIjiiSGMAe1bwEuIHrhY,10893
265
- ipex_llm-2.2.0b20250324.data/scripts/ipex-llm-init,sha256=fLQsT2dRL6H5bThb4GuIWotAuqoLsIxFwA-0c2qmaO8,6672
266
- ipex_llm-2.2.0b20250324.data/scripts/llm-chat,sha256=TdUnUmNapzuoe1c8IzrdVOQwWEg8IqsMSBRlOD3daZM,2249
267
- ipex_llm-2.2.0b20250324.data/scripts/llm-cli,sha256=RXGPlLElHxcKzoUxljEMBIAXbzCDysXL-Nxw-xF-7LU,2457
268
- ipex_llm-2.2.0b20250324.dist-info/METADATA,sha256=-08H2IVB6tKDeY2wZAIa6okFhO-0bKSiPAT4UEgVffs,13917
269
- ipex_llm-2.2.0b20250324.dist-info/WHEEL,sha256=PPJcBMAZibF_2GFE9NmOJGqiaSMPiNFbJd6QaJjdA6Y,109
270
- ipex_llm-2.2.0b20250324.dist-info/entry_points.txt,sha256=TiUyBB2MRmfF3ko-pyAEzqeBCRnyhu27bNOAsWPp3e8,61
271
- ipex_llm-2.2.0b20250324.dist-info/top_level.txt,sha256=CGCMHM-SyqUabU4h8RqJ2KTYckQUO3LvIWwmUQ6Qbzw,9
272
- ipex_llm-2.2.0b20250324.dist-info/RECORD,,
265
+ ipex_llm-2.2.0b20250326.data/scripts/ipex-llm-init,sha256=fLQsT2dRL6H5bThb4GuIWotAuqoLsIxFwA-0c2qmaO8,6672
266
+ ipex_llm-2.2.0b20250326.data/scripts/llm-chat,sha256=TdUnUmNapzuoe1c8IzrdVOQwWEg8IqsMSBRlOD3daZM,2249
267
+ ipex_llm-2.2.0b20250326.data/scripts/llm-cli,sha256=RXGPlLElHxcKzoUxljEMBIAXbzCDysXL-Nxw-xF-7LU,2457
268
+ ipex_llm-2.2.0b20250326.dist-info/METADATA,sha256=fHSU2c7G6TDPCkVikxp0pUvb-oq-QgVvbFVE1Ea17BM,13917
269
+ ipex_llm-2.2.0b20250326.dist-info/WHEEL,sha256=PPJcBMAZibF_2GFE9NmOJGqiaSMPiNFbJd6QaJjdA6Y,109
270
+ ipex_llm-2.2.0b20250326.dist-info/entry_points.txt,sha256=TiUyBB2MRmfF3ko-pyAEzqeBCRnyhu27bNOAsWPp3e8,61
271
+ ipex_llm-2.2.0b20250326.dist-info/top_level.txt,sha256=CGCMHM-SyqUabU4h8RqJ2KTYckQUO3LvIWwmUQ6Qbzw,9
272
+ ipex_llm-2.2.0b20250326.dist-info/RECORD,,