ipex-llm 2.3.0b20250513__py3-none-manylinux2010_x86_64.whl → 2.3.0b20250515__py3-none-manylinux2010_x86_64.whl

ipex_llm/serving/fastchat/model_worker.py

@@ -470,7 +470,7 @@ if __name__ == "__main__":
 
     if args.gpus:
         invalidInputError(len(args.gpus.split(",")) > args.num_gpus, f"Larger --num-gpus "
-                          "({args.num_gpus}) than --gpus {args.gpus}!")
+                          f"({args.num_gpus}) than --gpus {args.gpus}!")
         os.environ["CUDA_VISIBLE_DEVICES"] = args.gpus
 
     gptq_config = GptqConfig(

ipex_llm/transformers/model.py

@@ -672,7 +672,7 @@ class _BaseAutoModelClass:
                 else:
                     invalidInputError(False,
                                       f'`torch_dtype` can be either `torch.dtype` or `"auto"`,'
-                                      'but received {torch_dtype}')
+                                      f'but received {torch_dtype}')
             dtype_orig = model_class._set_default_torch_dtype(torch_dtype)
 
         # Pretrained Model

ipex_llm/transformers/npu_model.py

@@ -217,7 +217,7 @@ class _BaseAutoModelClass:
                     max_prompt_len < max_context_len,
                     (
                         f"max_prompt_len ({max_prompt_len}) should be less"
-                        " than max_context_len ({max_context_len})"
+                        f" than max_context_len ({max_context_len})"
                     ),
                 )
                 optimize_kwargs = {
@@ -553,7 +553,7 @@ class _BaseAutoModelClass:
                     invalidInputError(
                         False,
                         f'`torch_dtype` can be either `torch.dtype` or `"auto"`,'
-                        "but received {torch_dtype}",
+                        f"but received {torch_dtype}",
                     )
             dtype_orig = model_class._set_default_torch_dtype(torch_dtype)
 
@@ -588,7 +588,7 @@ class _BaseAutoModelClass:
                 max_prompt_len < max_context_len,
                 (
                     f"max_prompt_len ({max_prompt_len}) should be less"
-                    " than max_context_len ({max_context_len})"
+                    f" than max_context_len ({max_context_len})"
                 ),
             )
             from ipex_llm.transformers.npu_models.convert_mp import optimize_llm_pre

ipex_llm/transformers/npu_models/phi3.py

@@ -127,7 +127,7 @@ def phi3_attention_forward(
                 invalidInputError(
                     False,
                     f"Attention mask should be of size {(bsz, 1, q_len, kv_seq_len)},"
-                    " but is {attention_mask.size()}"
+                    f" but is {attention_mask.size()}"
                 )
             attn_weights = attn_weights + attention_mask
 

ipex_llm/transformers/speculative.py

@@ -54,7 +54,8 @@ if version.parse(trans_version) >= version.parse("4.39.0"):
     try:
         from trl.core import top_k_top_p_filtering
     except ModuleNotFoundError:
-        log4Error.invalidInputError(False, "For transformers version >= 4.39.0, pip install trl")
+        log4Error.invalidInputError(False,
+                                    "For transformers version >= 4.39.0, pip install trl==0.11.0")
 else:
     from transformers import top_k_top_p_filtering
 

ipex_llm/transformers/utils.py

@@ -92,7 +92,7 @@ def load_state_dict(checkpoint_file: Union[str, os.PathLike]):
     except Exception as e:
         invalidInputError(False,
                           f"Unable to load weights"
-                          "from pytorch checkpoint file for '{checkpoint_file}' "
+                          f"from pytorch checkpoint file for '{checkpoint_file}' "
                           f"at '{checkpoint_file}'. ")
 
 

ipex_llm/utils/lazy_load_torch.py

@@ -112,7 +112,7 @@ def _load(pickle_fp, map_location, picklemoudle, pickle_file='data.pkl', zip_fil
                 data = fp.read(size)
                 return torch.frombuffer(bytearray(data), dtype=dtype)
             description = f'storage data_type={data_type} ' \
-                          'path-in-zip={filename} path={self.zip_file.filename}'
+                          f'path-in-zip={filename} path={self.zip_file.filename}'
             return LazyStorage(load=load, kind=pid[1], description=description)
 
         @staticmethod

{ipex_llm-2.3.0b20250513.dist-info → ipex_llm-2.3.0b20250515.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: ipex-llm
-Version: 2.3.0b20250513
+Version: 2.3.0b20250515
 Summary: Large Language Model Develop Toolkit
 Home-page: https://github.com/intel-analytics/ipex-llm
 Author: BigDL Authors
@@ -27,7 +27,7 @@ Requires-Dist: intel-openmp ; (platform_machine == "x86_64" or platform_machine
 Requires-Dist: torch ==2.1.2+cpu ; (platform_system == "Linux") and extra == 'all'
 Requires-Dist: torch ==2.1.2 ; (platform_system == "Windows") and extra == 'all'
 Provides-Extra: cpp
-Requires-Dist: bigdl-core-cpp ==2.7.0b20250513 ; extra == 'cpp'
+Requires-Dist: bigdl-core-cpp ==2.7.0b20250515 ; extra == 'cpp'
 Requires-Dist: setuptools ; extra == 'cpp'
 Requires-Dist: onednn-devel ==2025.0.1 ; (platform_system == "Windows") and extra == 'cpp'
 Requires-Dist: onednn ==2025.0.1 ; (platform_system == "Windows") and extra == 'cpp'
@@ -60,7 +60,7 @@ Requires-Dist: transformers ==4.40.0 ; extra == 'npu'
 Requires-Dist: intel-openmp ; (platform_machine == "x86_64" or platform_machine == "AMD64") and extra == 'npu'
 Requires-Dist: torch ==2.1.2+cpu ; (platform_system == "Linux") and extra == 'npu'
 Requires-Dist: torch ==2.1.2 ; (platform_system == "Windows") and extra == 'npu'
-Requires-Dist: bigdl-core-npu ==2.7.0b20250513 ; (platform_system == "Windows") and extra == 'npu'
+Requires-Dist: bigdl-core-npu ==2.7.0b20250515 ; (platform_system == "Windows") and extra == 'npu'
 Provides-Extra: serving
 Requires-Dist: py-cpuinfo ; extra == 'serving'
 Requires-Dist: fschat[model_worker,webui] ==0.2.36 ; extra == 'serving'
@@ -80,9 +80,9 @@ Requires-Dist: setuptools <70.0.0 ; extra == 'xpu'
 Requires-Dist: torch ==2.1.0a0 ; extra == 'xpu'
 Requires-Dist: torchvision ==0.16.0a0 ; extra == 'xpu'
 Requires-Dist: intel-extension-for-pytorch ==2.1.10+xpu ; extra == 'xpu'
-Requires-Dist: bigdl-core-xe-21 ==2.7.0b20250513 ; extra == 'xpu'
-Requires-Dist: bigdl-core-xe-batch-21 ==2.7.0b20250513 ; extra == 'xpu'
-Requires-Dist: bigdl-core-xe-addons-21 ==2.7.0b20250513 ; extra == 'xpu'
+Requires-Dist: bigdl-core-xe-21 ==2.7.0b20250515 ; extra == 'xpu'
+Requires-Dist: bigdl-core-xe-batch-21 ==2.7.0b20250515 ; extra == 'xpu'
+Requires-Dist: bigdl-core-xe-addons-21 ==2.7.0b20250515 ; extra == 'xpu'
 Provides-Extra: xpu-2-1
 Requires-Dist: py-cpuinfo ; extra == 'xpu-2-1'
 Requires-Dist: protobuf ; extra == 'xpu-2-1'
@@ -97,9 +97,9 @@ Requires-Dist: setuptools <70.0.0 ; extra == 'xpu-2-1'
 Requires-Dist: torch ==2.1.0a0 ; extra == 'xpu-2-1'
 Requires-Dist: torchvision ==0.16.0a0 ; extra == 'xpu-2-1'
 Requires-Dist: intel-extension-for-pytorch ==2.1.10+xpu ; extra == 'xpu-2-1'
-Requires-Dist: bigdl-core-xe-21 ==2.7.0b20250513 ; extra == 'xpu-2-1'
-Requires-Dist: bigdl-core-xe-batch-21 ==2.7.0b20250513 ; extra == 'xpu-2-1'
-Requires-Dist: bigdl-core-xe-addons-21 ==2.7.0b20250513 ; extra == 'xpu-2-1'
+Requires-Dist: bigdl-core-xe-21 ==2.7.0b20250515 ; extra == 'xpu-2-1'
+Requires-Dist: bigdl-core-xe-batch-21 ==2.7.0b20250515 ; extra == 'xpu-2-1'
+Requires-Dist: bigdl-core-xe-addons-21 ==2.7.0b20250515 ; extra == 'xpu-2-1'
 Requires-Dist: intel-openmp ; (platform_machine == "x86_64" or platform_machine == "AMD64") and extra == 'xpu-2-1'
 Requires-Dist: dpcpp-cpp-rt ==2024.0.2 ; (platform_system == "Windows") and extra == 'xpu-2-1'
 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'
 Requires-Dist: torch ==2.6.0+xpu ; extra == 'xpu-2-6'
 Requires-Dist: torchvision ==0.21.0+xpu ; extra == 'xpu-2-6'
 Requires-Dist: torchaudio ==2.6.0+xpu ; extra == 'xpu-2-6'
-Requires-Dist: bigdl-core-xe-all ==2.7.0b20250513 ; extra == 'xpu-2-6'
+Requires-Dist: bigdl-core-xe-all ==2.7.0b20250515 ; extra == 'xpu-2-6'
 Requires-Dist: onednn-devel ==2025.0.1 ; extra == 'xpu-2-6'
 Requires-Dist: onednn ==2025.0.1 ; extra == 'xpu-2-6'
 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'
 Requires-Dist: accelerate ==0.23.0 ; extra == 'xpu-2-6-arl'
 Requires-Dist: tabulate ; extra == 'xpu-2-6-arl'
 Requires-Dist: setuptools ; extra == 'xpu-2-6-arl'
-Requires-Dist: bigdl-core-xe-all ==2.7.0b20250513 ; extra == 'xpu-2-6-arl'
+Requires-Dist: bigdl-core-xe-all ==2.7.0b20250515 ; extra == 'xpu-2-6-arl'
 Requires-Dist: onednn-devel ==2025.0.1 ; extra == 'xpu-2-6-arl'
 Requires-Dist: onednn ==2025.0.1 ; extra == 'xpu-2-6-arl'
 Requires-Dist: dpcpp-cpp-rt ==2025.0.2 ; extra == 'xpu-2-6-arl'

{ipex_llm-2.3.0b20250513.dist-info → ipex_llm-2.3.0b20250515.dist-info}/RECORD

@@ -89,7 +89,7 @@ ipex_llm/serving/fastapi/tgi_protocol.py,sha256=rPySgmBODkJ14OrS_KJaFe5RpqkHNH9c
 ipex_llm/serving/fastchat/__init__.py,sha256=b2IXvVqQ5cItki021h8s3ymW12RPu8QNPprq4Mn3bDM,586
 ipex_llm/serving/fastchat/bigdl_llm_model.py,sha256=NXEN_3EPmcP3dDnvug4MokEXXE2zVUnENgBYxfubqic,10084
 ipex_llm/serving/fastchat/ipex_llm_worker.py,sha256=vCqPi_ISuWU02Ngs6nEihwW-e8CDyazQa6qsXv-hvLU,19647
-ipex_llm/serving/fastchat/model_worker.py,sha256=qJSLyWNkP6z70ysq4AV5SqHzXPJJiX2tz7AORB6jEvM,16649
+ipex_llm/serving/fastchat/model_worker.py,sha256=jvRrrSs8rDQXR1Q4J8cq6VeBozU-TW-zaAerrLdkQHc,16650
 ipex_llm/serving/fastchat/tgi_api_protocol.py,sha256=brT3k3-V0NJrU4fRqUwWjC0O3iOitdttDfduXXEefh0,5918
 ipex_llm/serving/fastchat/tgi_api_server.py,sha256=agNTAEiZPSuj3dEdIdYKwkoY0cXOUDX06DiM9VP2knQ,24418
 ipex_llm/serving/fastchat/vllm_worker.py,sha256=ZLz2Q9GxJO6r_LOiP6epgCRjBGk-K4EB1SNEWSJp5DA,11091
@@ -102,17 +102,17 @@ ipex_llm/transformers/lisa.py,sha256=F5WxbtXQ7RdKulj83h_2DnEIgKiKGZf7zvOmg6QBl2s
 ipex_llm/transformers/loader.py,sha256=c9qfJSC6-in-mkd-iKb1igk3nHWUYS3QtyH2cOazmKc,6825
 ipex_llm/transformers/lookup.py,sha256=b6OlZ9OV10R9qeWw8mVryVpDxszkjwLkldvi7GPMJY8,19614
 ipex_llm/transformers/low_bit_linear.py,sha256=1S8H684odAx5ZVDx_qNMv4FFrkJekZduao8datZYiqw,39201
-ipex_llm/transformers/model.py,sha256=r9jLsLlGB4rOZ7SMrmCE5qyXqC9Ae56unNq1gNjfP54,40885
+ipex_llm/transformers/model.py,sha256=tWTzKsCz8A1P5gYEeG9KZgpxQgbP9hQ-TWAdkebA6Jg,40886
 ipex_llm/transformers/modelling_bigdl.py,sha256=7JpNVMuyq_OmtNUaMFMXdxPWZp2q0QHC02QeA-VTPOw,6709
-ipex_llm/transformers/npu_model.py,sha256=zgXOiLIJ-3p-1Kejgv4jUFK8OiBZbezMZrRyn0_6_8c,40306
+ipex_llm/transformers/npu_model.py,sha256=X8mdY6N9TYlxG41wmFloX44ZUjyitFzdKbhzO7TToFY,40309
 ipex_llm/transformers/patches.py,sha256=G9KcXxo42H1HJEDaroq4JbBN5P0P0lty7U7kk7-g4tw,991
 ipex_llm/transformers/pipeline_parallel.py,sha256=uNZpOXljNmdoEYnP8U-VFiN4dRZb2piQbIf2bG9LQnE,49051
 ipex_llm/transformers/qlora.py,sha256=qV9Y6G5kAaet77LLA3oXn3qQY4ayyAPZ7NAjOlHCS7g,14967
 ipex_llm/transformers/relora.py,sha256=-dYzUV0P-IhO2jFdnzN9-v_sFzJpRj3ZwN9eCJzOoCw,16567
-ipex_llm/transformers/speculative.py,sha256=0XNLgc9dGswJHVPrXo4iM7pPxkWwfFfJMECcivJSnIc,63368
+ipex_llm/transformers/speculative.py,sha256=_5LkyLvdAtkpMfOFBhzbtgHS7MzkJz_EbNpTt4gLQEg,63412
 ipex_llm/transformers/streamer.py,sha256=RrVlLblzCOtABRUpaMXAyaMnCGgLUtAi_YesLumRbww,4842
 ipex_llm/transformers/training_patch.py,sha256=oxMkUtqyvqJiprw6dE3skkYfD1HOmUlH9N0hBkbn0G0,10799
-ipex_llm/transformers/utils.py,sha256=a-2wbflSd_yYnC5qcMoY5HLR1yT_QpxeX_WpGpaDLrA,17457
+ipex_llm/transformers/utils.py,sha256=FCin1mKRmDyQ9J8dCJgcPOhOh1DUtawrL0wZZ7iSnOw,17458
 ipex_llm/transformers/xgrammar.py,sha256=dd-e0DO0s-t-idngVzncnPAa_Gxb8YNoRJ3WROkwrs8,1840
 ipex_llm/transformers/xpu_customize_fwd.py,sha256=PUBYLnTbaBXUs3Dnte9Gqln2XFk8iA62SmloWjr7GJI,7668
 ipex_llm/transformers/xpu_ops.py,sha256=z95iTtcDQvNyJOvB4A6B_ECTYjHp4A7x-FsssoETOMs,4914
@@ -214,7 +214,7 @@ ipex_llm/transformers/npu_models/mistral.py,sha256=iRdmIQI_bbbZxRCYRvnV4rWjX2t-6
 ipex_llm/transformers/npu_models/mp_models_base.py,sha256=19iXZhX7SBQNXlAANn4gB1Enrk1FDmsGkmgI0m19vKg,28555
 ipex_llm/transformers/npu_models/npu_llm_cpp.py,sha256=B40sBujvy31ETFBgcYAf4CN23UuTCBEJVaxjIMaoEHk,4268
 ipex_llm/transformers/npu_models/paraformer_mp.py,sha256=lGEjmKHW_Pk3BE3nqa1ZVgJ3P5p4lNp7p6wMV7KrtCU,37871
-ipex_llm/transformers/npu_models/phi3.py,sha256=R-EuqHsTrPTX33HtCGAMFlRdXB_j5mH_7FDnj62JtNM,6555
+ipex_llm/transformers/npu_models/phi3.py,sha256=n3WRQMcaL1iJOuchQh3P-MCulSCYWVzNM_JCh32q3_o,6556
 ipex_llm/transformers/npu_models/phi3_v.py,sha256=EMZuTPkGfuDVp9c5BU1HyzXHWKswHRQ8bvQjzocIyHA,7737
 ipex_llm/transformers/npu_models/qwen2.py,sha256=RDNtPK8kxMk3z8A4S53saTrw2klgkzo4oa7voJLwr1o,12085
 ipex_llm/transformers/npu_models/qwen2_mp.py,sha256=EKiI80rnQ43WUF_2wWCy75mx-rbjAbRQSB49OgjZFNo,45003
@@ -241,7 +241,7 @@ ipex_llm/utils/convert_util.py,sha256=X1eLTdF9crlca6jPEQKymEITRx_57i_87dhl1htPEh
 ipex_llm/utils/glibc_checker.py,sha256=bm6kN6gbpA7GKtnUgsCE6K16iZZyil-Ylp55SoRvHG8,2093
 ipex_llm/utils/ipex_importer.py,sha256=pQu3IbvXZjgSEw7T42EZwbfVXpCYEm5tB2QHtzYmp1I,5922
 ipex_llm/utils/isa_checker.py,sha256=SisZ32B1G7meZvgRKZcani1WevIhMwum7nilZ3sHgXY,2259
-ipex_llm/utils/lazy_load_torch.py,sha256=MCZZr1NWwxOZQt-3OthNBxMxdN2jiRbvyXwJkrIKAhM,7129
+ipex_llm/utils/lazy_load_torch.py,sha256=GeolOQFIj2Lnn47nUCGx66bqEqznW0my4atbTQJp6Us,7130
 ipex_llm/utils/modules.py,sha256=LZ7FvggSSb2QMHYeeHDmtI8ziIvXVHsF4hV-vOUK60Q,764
 ipex_llm/utils/utils.py,sha256=jfDRc622QF2U2YZpmqo0evnaElHbVBDoRZqH_6hUXmI,1556
 ipex_llm/utils/common/__init__.py,sha256=pV_8I-oa8RrHSCJg6ADBY_Tre6vXCqAMyhfpj1UhvYE,988
@@ -265,11 +265,11 @@ ipex_llm/vllm/xpu/engine/__init__.py,sha256=sOvwLx_Zj0jiRCGj9W3DgGTfcSU3hABYhgIQ
 ipex_llm/vllm/xpu/engine/engine.py,sha256=XAprw7VifjfnR915TZOaKcxe3QCFsVBgxzS8qOdn1yg,14462
 ipex_llm/vllm/xpu/entrypoints/openai/api_server.py,sha256=uWHyyHLw-B8wXBnQw9_MCG81tKK9Jb0dyq1xfYHgoNw,45905
 ipex_llm/vllm/xpu/entrypoints/openai/cli_args.py,sha256=hB398yYtKauASRzevctScdbFIjiiSGMAe1bwEuIHrhY,10893
-ipex_llm-2.3.0b20250513.data/scripts/ipex-llm-init,sha256=fLQsT2dRL6H5bThb4GuIWotAuqoLsIxFwA-0c2qmaO8,6672
-ipex_llm-2.3.0b20250513.data/scripts/llm-chat,sha256=TdUnUmNapzuoe1c8IzrdVOQwWEg8IqsMSBRlOD3daZM,2249
-ipex_llm-2.3.0b20250513.data/scripts/llm-cli,sha256=RXGPlLElHxcKzoUxljEMBIAXbzCDysXL-Nxw-xF-7LU,2457
-ipex_llm-2.3.0b20250513.dist-info/METADATA,sha256=9uv-v2lumKlA4hk3TSWqEyefcf1w438ypWpN5hEZdMM,8865
-ipex_llm-2.3.0b20250513.dist-info/WHEEL,sha256=PPJcBMAZibF_2GFE9NmOJGqiaSMPiNFbJd6QaJjdA6Y,109
-ipex_llm-2.3.0b20250513.dist-info/entry_points.txt,sha256=TiUyBB2MRmfF3ko-pyAEzqeBCRnyhu27bNOAsWPp3e8,61
-ipex_llm-2.3.0b20250513.dist-info/top_level.txt,sha256=CGCMHM-SyqUabU4h8RqJ2KTYckQUO3LvIWwmUQ6Qbzw,9
-ipex_llm-2.3.0b20250513.dist-info/RECORD,,
+ipex_llm-2.3.0b20250515.data/scripts/ipex-llm-init,sha256=fLQsT2dRL6H5bThb4GuIWotAuqoLsIxFwA-0c2qmaO8,6672
+ipex_llm-2.3.0b20250515.data/scripts/llm-chat,sha256=TdUnUmNapzuoe1c8IzrdVOQwWEg8IqsMSBRlOD3daZM,2249
+ipex_llm-2.3.0b20250515.data/scripts/llm-cli,sha256=RXGPlLElHxcKzoUxljEMBIAXbzCDysXL-Nxw-xF-7LU,2457
+ipex_llm-2.3.0b20250515.dist-info/METADATA,sha256=_lbNuaDcLJThqZdxaHamNmc-qDmC7S7F7E_MgDqH_lk,8865
+ipex_llm-2.3.0b20250515.dist-info/WHEEL,sha256=PPJcBMAZibF_2GFE9NmOJGqiaSMPiNFbJd6QaJjdA6Y,109
+ipex_llm-2.3.0b20250515.dist-info/entry_points.txt,sha256=TiUyBB2MRmfF3ko-pyAEzqeBCRnyhu27bNOAsWPp3e8,61
+ipex_llm-2.3.0b20250515.dist-info/top_level.txt,sha256=CGCMHM-SyqUabU4h8RqJ2KTYckQUO3LvIWwmUQ6Qbzw,9
+ipex_llm-2.3.0b20250515.dist-info/RECORD,,