ipex-llm 2.3.0b20250414__py3-none-win_amd64.whl → 2.3.0b20250417__py3-none-win_amd64.whl

ipex_llm/transformers/models/baichuan.py

@@ -326,14 +326,17 @@ def baichuan_attention_forward_13b(
         else:
             attention_mask = attention_mask[None, :, -q_len:, :]
 
+    head_dim = query_states.shape[-1]
+    scale = 1 / math.sqrt(head_dim)
+
     if use_sdp(q_len, kv_seq_len, self.head_dim, query_states):
         import xe_addons
         if use_quantize_kv:
             attn_output = xe_addons.sdp_fp8(query_states, key_states, value_states,
-                                            attention_mask)
+                                            attention_mask, scale)
         else:
             attn_output = xe_addons.sdp(query_states, key_states, value_states,
-                                        attention_mask)
+                                        attention_mask, scale)
         attn_weights = None
     else:
         if use_quantize_kv:

ipex_llm/transformers/models/chatglm.py

@@ -68,7 +68,9 @@ def glm_sdpa(query, key, value, attention_mask=None, is_causal=False):
         if use_sdp(query.shape[2], key.shape[2],
                    query.shape[-1], query):
             import xe_addons
-            attn_output = xe_addons.sdp(query, key, value, attn_bias)
+            head_dim = query.shape[-1]
+            scale = 1 / math.sqrt(head_dim)
+            attn_output = xe_addons.sdp(query, key, value, attn_bias, scale)
             context_layer = attn_output.view(query.shape)
         else:
             head_dim = query.size(-1)

ipex_llm/transformers/models/qwen_vl.py

@@ -164,7 +164,9 @@ def qwen_attention_forward_vl(
     if not self.training and not hidden_states.requires_grad and \
             use_sdp(q_len, key.shape[2], self.head_dim, query):
         import xe_addons
-        attn_output = xe_addons.sdp(query, key, value, attention_mask)
+        head_dim = query.shape[-1]
+        scale = 1 / math.sqrt(head_dim)
+        attn_output = xe_addons.sdp(query, key, value, attention_mask, scale)
         attn_output = attn_output.view(query.shape)
         attn_output = attn_output.transpose(1, 2)
         attn_weight = None

{ipex_llm-2.3.0b20250414.dist-info → ipex_llm-2.3.0b20250417.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: ipex-llm
-Version: 2.3.0b20250414
+Version: 2.3.0b20250417
 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 ==2.7.0b20250414 ; extra == 'cpp'
+Requires-Dist: bigdl-core-cpp ==2.7.0b20250417 ; 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 ==2.7.0b20250414 ; (platform_system == "Windows") and extra == 'npu'
+Requires-Dist: bigdl-core-npu ==2.7.0b20250417 ; (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 ==2.7.0b20250414 ; extra == 'xpu'
-Requires-Dist: bigdl-core-xe-batch-21 ==2.7.0b20250414 ; extra == 'xpu'
-Requires-Dist: bigdl-core-xe-addons-21 ==2.7.0b20250414 ; extra == 'xpu'
+Requires-Dist: bigdl-core-xe-21 ==2.7.0b20250417 ; extra == 'xpu'
+Requires-Dist: bigdl-core-xe-batch-21 ==2.7.0b20250417 ; extra == 'xpu'
+Requires-Dist: bigdl-core-xe-addons-21 ==2.7.0b20250417 ; 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 ==2.7.0b20250414 ; extra == 'xpu-2-1'
-Requires-Dist: bigdl-core-xe-batch-21 ==2.7.0b20250414 ; extra == 'xpu-2-1'
-Requires-Dist: bigdl-core-xe-addons-21 ==2.7.0b20250414 ; extra == 'xpu-2-1'
+Requires-Dist: bigdl-core-xe-21 ==2.7.0b20250417 ; extra == 'xpu-2-1'
+Requires-Dist: bigdl-core-xe-batch-21 ==2.7.0b20250417 ; extra == 'xpu-2-1'
+Requires-Dist: bigdl-core-xe-addons-21 ==2.7.0b20250417 ; 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 ==2.7.0b20250414 ; extra == 'xpu-2-6'
+Requires-Dist: bigdl-core-xe-all ==2.7.0b20250417 ; 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 ==2.7.0b20250414 ; extra == 'xpu-2-6-arl'
+Requires-Dist: bigdl-core-xe-all ==2.7.0b20250417 ; 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'
@@ -155,9 +155,9 @@ Requires-Dist: tokenizers ==0.15.2 ; extra == 'xpu-arc'
 Requires-Dist: accelerate ==0.23.0 ; extra == 'xpu-arc'
 Requires-Dist: tabulate ; extra == 'xpu-arc'
 Requires-Dist: setuptools ; extra == 'xpu-arc'
-Requires-Dist: bigdl-core-xe-23 ==2.7.0b20250414 ; extra == 'xpu-arc'
-Requires-Dist: bigdl-core-xe-batch-23 ==2.7.0b20250414 ; extra == 'xpu-arc'
-Requires-Dist: bigdl-core-xe-addons-23 ==2.7.0b20250414 ; extra == 'xpu-arc'
+Requires-Dist: bigdl-core-xe-23 ==2.7.0b20250417 ; extra == 'xpu-arc'
+Requires-Dist: bigdl-core-xe-batch-23 ==2.7.0b20250417 ; extra == 'xpu-arc'
+Requires-Dist: bigdl-core-xe-addons-23 ==2.7.0b20250417 ; extra == 'xpu-arc'
 Requires-Dist: intel-openmp ; (platform_machine == "x86_64" or platform_machine == "AMD64") and extra == 'xpu-arc'
 Requires-Dist: torch ==2.3.1+cxx11.abi ; (platform_system == "Linux") and extra == 'xpu-arc'
 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'
 Requires-Dist: accelerate ==0.23.0 ; extra == 'xpu-arl'
 Requires-Dist: tabulate ; extra == 'xpu-arl'
 Requires-Dist: setuptools ; extra == 'xpu-arl'
-Requires-Dist: bigdl-core-xe-23 ==2.7.0b20250414 ; extra == 'xpu-arl'
-Requires-Dist: bigdl-core-xe-batch-23 ==2.7.0b20250414 ; extra == 'xpu-arl'
-Requires-Dist: bigdl-core-xe-addons-23 ==2.7.0b20250414 ; extra == 'xpu-arl'
+Requires-Dist: bigdl-core-xe-23 ==2.7.0b20250417 ; extra == 'xpu-arl'
+Requires-Dist: bigdl-core-xe-batch-23 ==2.7.0b20250417 ; extra == 'xpu-arl'
+Requires-Dist: bigdl-core-xe-addons-23 ==2.7.0b20250417 ; extra == 'xpu-arl'
 Requires-Dist: intel-openmp ; (platform_machine == "x86_64" or platform_machine == "AMD64") and extra == 'xpu-arl'
 Requires-Dist: torch ==2.3.1+cxx11.abi ; (platform_system == "Linux") and extra == 'xpu-arl'
 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'
 Requires-Dist: accelerate ==0.23.0 ; extra == 'xpu-lnl'
 Requires-Dist: tabulate ; extra == 'xpu-lnl'
 Requires-Dist: setuptools ; extra == 'xpu-lnl'
-Requires-Dist: bigdl-core-xe-23 ==2.7.0b20250414 ; extra == 'xpu-lnl'
-Requires-Dist: bigdl-core-xe-batch-23 ==2.7.0b20250414 ; extra == 'xpu-lnl'
-Requires-Dist: bigdl-core-xe-addons-23 ==2.7.0b20250414 ; extra == 'xpu-lnl'
+Requires-Dist: bigdl-core-xe-23 ==2.7.0b20250417 ; extra == 'xpu-lnl'
+Requires-Dist: bigdl-core-xe-batch-23 ==2.7.0b20250417 ; extra == 'xpu-lnl'
+Requires-Dist: bigdl-core-xe-addons-23 ==2.7.0b20250417 ; extra == 'xpu-lnl'
 Requires-Dist: intel-openmp ; (platform_machine == "x86_64" or platform_machine == "AMD64") and extra == 'xpu-lnl'
 Requires-Dist: torch ==2.3.1+cxx11.abi ; (platform_system == "Linux") and extra == 'xpu-lnl'
 Requires-Dist: torchvision ==0.18.1+cxx11.abi ; (platform_system == "Linux") and extra == 'xpu-lnl'

{ipex_llm-2.3.0b20250414.dist-info → ipex_llm-2.3.0b20250417.dist-info}/RECORD

@@ -41,35 +41,35 @@ ipex_llm/langchain/llms/transformerspipelinellm.py,sha256=vm522YPPwWxxAPVvQBtxRf
 ipex_llm/langchain/vllm/__init__.py,sha256=T-EbRT6GJ_8RCu-iLmSzcftOimXSPQf2d5X72AUAy2Y,874
 ipex_llm/langchain/vllm/vllm.py,sha256=6dxc-ZISZQrJilEa_HA827l75Dv9rcHpY_G6FdJ8BVs,7793
 ipex_llm/libs/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-ipex_llm/libs/bloom-api.dll,sha256=RQH35fsCjbpJAqZ6vqpzIvurExHg77Dm_j9SF1eL8DQ,36352
-ipex_llm/libs/bloom.dll,sha256=riUIPG6kaX1CeCirSMMuPjnadxQrm89t3s8MCjW9VDg,507904
-ipex_llm/libs/gptneox-api.dll,sha256=3Z21zemTxGTPtZyLrJQlwaphhjnEUH_nD2_azE4GHjU,24576
-ipex_llm/libs/gptneox.dll,sha256=Y91HtjmiOHXOgttQod7KG9f9KkCdU9Jbu_V_2Ncz1Ic,568320
-ipex_llm/libs/libbloom_avx.dll,sha256=siYYEY3keJNXxD1q1qe2EcX0Cz6ezyr4E1jD4kQ8DMI,536576
-ipex_llm/libs/libbloom_vnni.dll,sha256=Q0Wtp6u-xcTW_TTsUHqn0X7353W_ere_k0GwsI0QVhE,508416
-ipex_llm/libs/libgptneox_avx.dll,sha256=AtpJopCdHu0RAVPyunHsQ9qQWLiJMIE7ba4gme7nC_k,596992
-ipex_llm/libs/libgptneox_vnni.dll,sha256=Gt4tFPro-zNdU4S22k96Tds-tVCwRgccuu3RRwWCsJ4,568832
-ipex_llm/libs/libllama_avx.dll,sha256=BrqmZm6ASzZ5hPiu7cjJKLxs2pnTUDS8vlaVwXtglKc,591360
-ipex_llm/libs/libllama_vnni.dll,sha256=44HGXEJgJYJ6kS_dngqYR9VBFXRiYEooxo4K_xLLvyA,563200
-ipex_llm/libs/libstarcoder_avx.dll,sha256=XkCQzKH7Yg3n1hLbYhNaCgRGa_vEjT4G048i7FqiPFI,627712
-ipex_llm/libs/libstarcoder_vnni.dll,sha256=hhggl9l81TYcEGpZymdRkAtGeyrgSUqdHR3Xn-yfARE,599552
-ipex_llm/libs/llama-api.dll,sha256=k50ow33IgEjCfmyhaeLEK77JRVhu8UacJGiVCdj56lI,25600
-ipex_llm/libs/llama.dll,sha256=c81XUKbRl5RRpvad8qji61I-TW2N2WIA7SPWTmJ6c7I,562688
-ipex_llm/libs/main-bloom.exe,sha256=-Pg9G61YLFlBuMloBWPZ0BwuyizTdP9Gmp9H8lBmlIE,103424
-ipex_llm/libs/main-gptneox.exe,sha256=FqJFdSuRLzHzISAS3Fq-hpNaD5AzByRhidVJGvx8XsM,98816
-ipex_llm/libs/main-llama.exe,sha256=gCMvDvxMGMcTITQVB7JwzDQl8hWiS7cMedpd5qH1Yso,99840
-ipex_llm/libs/main-starcoder.exe,sha256=Hp8bVAB3gkleaTILUFP373bj_-FfZBhgY98ifF4Seg8,157696
-ipex_llm/libs/pipeline.dll,sha256=klcRphRZj-Ver9bb-3YarFz1Aofx2eH_9ZFVj6005IY,73216
-ipex_llm/libs/quantize-bloom.exe,sha256=1RCyEGDHqMgEV3-Z5whNWeoPH2pbmIuoNdFxwwWVzBM,126464
-ipex_llm/libs/quantize-bloom_vnni.exe,sha256=9B3eXy6FuOxdw4R7226GWwZdYRkA_0wrrdoEB4iRvL4,128000
-ipex_llm/libs/quantize-gptneox.exe,sha256=ho-9GbDDXLe6OCZUQXLLKQBaxndRs0uoIp7blm7mkX4,104448
-ipex_llm/libs/quantize-gptneox_vnni.exe,sha256=YxBxpYZGZASjRE0AvfhWPFN5y07J7b_p7uQ6KBfcgoA,104960
-ipex_llm/libs/quantize-llama.exe,sha256=N4t0zFvqDydwEF-ziXUfiWtoBUU_0InDyBgufaUGXzU,110080
-ipex_llm/libs/quantize-llama_vnni.exe,sha256=23i9bbF_ezU4WLc2vSjdzPP2aTMdPyxiQMFnSAaLk1A,110592
-ipex_llm/libs/quantize-starcoder.exe,sha256=DYm5NLxvHgpeldjdK8e0LcvgnLauLo5gz78CGVEKQTI,127488
-ipex_llm/libs/quantize-starcoder_vnni.exe,sha256=uoMxJlhpV-BzhANKO6IJls6iIP7KG03HIq8KxZXp-A8,128512
-ipex_llm/libs/starcoder-api.dll,sha256=wKmuMWuK2rlCEPrSLIaaK_ylaECQEiU730vSyurxBG8,21504
-ipex_llm/libs/starcoder.dll,sha256=rWkn029EPDSsEjlRu18jLq_0fC4b6axEeHjgoezGSEM,599040
+ipex_llm/libs/bloom-api.dll,sha256=8GKkxBSPTcSyRjh157F7qeuY52IPayY0y6z02zrp9Ds,36352
+ipex_llm/libs/bloom.dll,sha256=40E1CsiikXT4AvJJJqhMdwshFKEu-bU5pGsfoaoxEtY,507904
+ipex_llm/libs/gptneox-api.dll,sha256=_i-r-OokErtb34pTYJldcKlNJXSVaXqTN0bYlHwGFI8,24576
+ipex_llm/libs/gptneox.dll,sha256=U-Nju_l5nso4Nav4bXe8sYqTN-1dNjZVIjYbCQSt4wQ,568320
+ipex_llm/libs/libbloom_avx.dll,sha256=x8xvB7OO94d_ETc5PDLJYn1ZnrYTiv22157gvXQbMVg,536576
+ipex_llm/libs/libbloom_vnni.dll,sha256=OCKzhd6VXduwrLhHCOJT2r5s53NQ4UtQNtAIFwXSzR8,508416
+ipex_llm/libs/libgptneox_avx.dll,sha256=trVw-EQfARG8ndy2KNvcCdYq0gR3bCoAx_Z7mvLFxO4,596992
+ipex_llm/libs/libgptneox_vnni.dll,sha256=Q8UWiGduZEVcEMUtWncXiIOmkcW4Sv19LyIEjS2Rf-w,568832
+ipex_llm/libs/libllama_avx.dll,sha256=3-6vo7hiAZTkmc95nDLsTJiic5bHb7k9Wcivf0l3dhQ,591360
+ipex_llm/libs/libllama_vnni.dll,sha256=GehXAWR7mtzqe_ZhBw7PrInR5ivasLkCmckC3xQdjCc,563200
+ipex_llm/libs/libstarcoder_avx.dll,sha256=gCnQAEKtiwl8blIxyGeYZD4TFFvgUh_uRTFaVZgKTzs,627712
+ipex_llm/libs/libstarcoder_vnni.dll,sha256=AscRUtIJ_pCCjmqThSQNwgLjg44sqR7IioK9Eq90_pE,599552
+ipex_llm/libs/llama-api.dll,sha256=_8qUadqe6BqctrenKreEGwKOKq3xJdNbzrlIuIY4CrE,25600
+ipex_llm/libs/llama.dll,sha256=jks2Rsf-mg9Da8J4QhrCr-bQ8q5hyCWazLtzJz8fq50,562688
+ipex_llm/libs/main-bloom.exe,sha256=8YTMY3qPokXroK6hQY5JKtghIcVFVqKY3nQhyxGgGP8,103424
+ipex_llm/libs/main-gptneox.exe,sha256=6u4YNL-v9RHOCvDU7hS6ru5B8oooXZLdtNaAW3i9s_A,98816
+ipex_llm/libs/main-llama.exe,sha256=wfCXNmQrmc9TC6FD_c7VKlYACUZM7-f0gh6LYc77IG0,99840
+ipex_llm/libs/main-starcoder.exe,sha256=s9ljwEiHVGnFtxXotrZkj2CaIyZ8mZVXDpmDq6HG_Ms,157696
+ipex_llm/libs/pipeline.dll,sha256=-WNZTxBZDQL4qSl5yMu31GNlH_wsE1kPF0-bb2vHb5Y,73216
+ipex_llm/libs/quantize-bloom.exe,sha256=dsBq5vIfxk4up_gvTZb_xfgKYokXp93oE5QNY_-kquo,126464
+ipex_llm/libs/quantize-bloom_vnni.exe,sha256=_m6sLNQBIwU_Tbgiti47VcHA_H37aRCshki4HxcoMAY,128000
+ipex_llm/libs/quantize-gptneox.exe,sha256=7ca9rJmsmejukxjYzB-dcFRAqndnnbVVWoHjSy_Fm8U,104448
+ipex_llm/libs/quantize-gptneox_vnni.exe,sha256=RP0WURiXe7hc6mjlKhNjmRuvgvyFsutwbklxUl1ICVU,104960
+ipex_llm/libs/quantize-llama.exe,sha256=gCoBeBfbTCwbE5ZnLofE0k6-MNtzjC2BHNg-guuAgp4,110080
+ipex_llm/libs/quantize-llama_vnni.exe,sha256=pbgZjdS5_53o6rJrjJ43WCzPx-Q_kQsrQNUU329qbbM,110592
+ipex_llm/libs/quantize-starcoder.exe,sha256=mrlqQoZvIX9UHGHIbtHeH_FT8bRxe2AC4tmM1VmZ3SU,127488
+ipex_llm/libs/quantize-starcoder_vnni.exe,sha256=rT4HXKlcMKS0xgCOASGNFt1hGLttbLc3GXi-RlCyFMs,128512
+ipex_llm/libs/starcoder-api.dll,sha256=4x_33RngI5Kgo8OA_cR4N2xgvb2d5bz2bifxkpveUaI,21504
+ipex_llm/libs/starcoder.dll,sha256=gXOKiNobl6oZcQ21_KUO68GAqwEcfw7CvdcokfF1V3g,599040
 ipex_llm/llamaindex/__init__.py,sha256=T-EbRT6GJ_8RCu-iLmSzcftOimXSPQf2d5X72AUAy2Y,874
 ipex_llm/llamaindex/llms/__init__.py,sha256=KP1lEdGqDuxPoxL1ZSH25Pm2kKMPJBWUTLR0ckSLMIU,1139
 ipex_llm/llamaindex/llms/bigdlllm.py,sha256=FQBzq1KOjfc6uofTXAha3O7TqpJkNfOFepXQmOVlbnI,26314
@@ -138,11 +138,11 @@ ipex_llm/transformers/gguf/models/model_implement/yuan2/configuration_yuan.py,sh
 ipex_llm/transformers/gguf/models/model_implement/yuan2/yuan_hf_model.py,sha256=_AOGMV65XHxgTxIib7lgs49InopcecTzRwgtYR8NTUg,51084
 ipex_llm/transformers/models/__init__.py,sha256=tp2DcVkKg1-QvdYk7DY7rZvQWCDQ4ZjU8NAQ7Fclrpg,584
 ipex_llm/transformers/models/aquila.py,sha256=VZb5Drpo_fTxwcExZ397LygnsNPX2sVbie9_JeFudZI,5252
-ipex_llm/transformers/models/baichuan.py,sha256=8b43mBRZJEf_xLNoodhA4r9x1anqwC3Wt8awWel-aUo,18306
+ipex_llm/transformers/models/baichuan.py,sha256=nRqg_ih8TCIAeSjoiQpaGNAXqXGpVvQmqj5_6gr-Rxg,18395
 ipex_llm/transformers/models/baichuan_m1.py,sha256=l6BD9jbA8TQ5Q0fcngHNRw73SxqvC126ErENs7Zh-do,9513
 ipex_llm/transformers/models/bert.py,sha256=U3JVnuY1z0ILlJgMvetOKuCAkH74ufGbg4whZ8GNtmQ,5620
 ipex_llm/transformers/models/bloom.py,sha256=PxfzyYT-nFn3K5rZhTQjmcEjUUzAhUFzxIN4kzRlCuc,8103
-ipex_llm/transformers/models/chatglm.py,sha256=UHai1t2AUtGmF765_eHF8LUMVQzp_oCBx8TJB21WrHk,12597
+ipex_llm/transformers/models/chatglm.py,sha256=DQM63oPIVMMTBQN4O4hPF4WY1aSiTWq4Bt0BPmgrK0k,12687
 ipex_llm/transformers/models/chatglm2.py,sha256=KyAIX7zGVQDQuwwM3QMBNWZbTeMHEzKUIgAryT0voHc,14933
 ipex_llm/transformers/models/chatglm4.py,sha256=QvUehdaCePB3MNHyWg3dneDxmjtBdxYeKUyQUVcsgfM,16886
 ipex_llm/transformers/models/chatglm4v.py,sha256=Ba9Xtzwtzk_rzg5khGqDrlHfJsDwc5YcM5_yPoord7o,13324
@@ -174,7 +174,7 @@ ipex_llm/transformers/models/qwen2.py,sha256=zK-FpUaxEhjD4gZa1ZvArodAilz29T_cpeA
 ipex_llm/transformers/models/qwen2_5_omni.py,sha256=gNnWvGiPntn6GlyIUtCVYCSIMPU6FG9TCS7fhWbwerY,18779
 ipex_llm/transformers/models/qwen2_moe.py,sha256=a0gYo-ngf8SxaEnBdZUJDnPS6Mkn_poDd8xqhx50icI,19516
 ipex_llm/transformers/models/qwen2_vl.py,sha256=G-9e2oN4f5p5IWQ-zsBZuONxTura3BjlgyT2meigbHQ,13579
-ipex_llm/transformers/models/qwen_vl.py,sha256=j7Nzzz2Qvynu9yrCXmoEfERjw43hXof5TbXIs7Ms-oY,17105
+ipex_llm/transformers/models/qwen_vl.py,sha256=lwNwCJNsBvOu1TGNDW_E2IlCjI7XZwHY8qIcZpuNDFc,17187
 ipex_llm/transformers/models/rwkv4.py,sha256=H4KMtxN0JA2ZTXnonHpsUUJ5xULemo-D1Jzl0ri_UY8,6123
 ipex_llm/transformers/models/rwkv5.py,sha256=OkRNj1pCAZg1z2Fw-I0DEnxLEdZyPeRSQ6msrkxLOCs,10710
 ipex_llm/transformers/models/sd.py,sha256=VvHV5u-0k2MgHu3NL9113hPj7DgfxqctuKzEEeNfRDU,5981
@@ -256,11 +256,11 @@ ipex_llm/vllm/xpu/engine/__init__.py,sha256=pY_CpyuZd72fr6s32ejeKHKFW0K4vUU2rzZj
 ipex_llm/vllm/xpu/engine/engine.py,sha256=NvCMbp0X8NVrOqbwm4FTvXOptTRLzu9jQsy37ZHnTk8,9493
 ipex_llm/vllm/xpu/entrypoints/openai/api_server.py,sha256=IjiSze9vzBCAkLu_VwIcJwuO1jyFna7DLrj6aSL7RaQ,35220
 ipex_llm/vllm/xpu/entrypoints/openai/cli_args.py,sha256=hB398yYtKauASRzevctScdbFIjiiSGMAe1bwEuIHrhY,10893
-ipex_llm-2.3.0b20250414.data/scripts/ipex-llm-init.bat,sha256=HPtCYuDYwEatq7dAwOvdfVcHYCpAVdbj75K1qh0vQek,2578
-ipex_llm-2.3.0b20250414.data/scripts/llm-chat.ps1,sha256=6qrs-hGVAV8IKh7Jx8nq_XrnZcjd7qGU5wndArM7Yag,2769
-ipex_llm-2.3.0b20250414.data/scripts/llm-cli.ps1,sha256=3qBtTLs_EjYDnM8YyCpJhzLnGCKTEGssu9UNqfkjVXs,3009
-ipex_llm-2.3.0b20250414.dist-info/METADATA,sha256=4kiLch6GAWOIkpZGeb-YiWgSUq8Linmj_qrwVY-hZew,13917
-ipex_llm-2.3.0b20250414.dist-info/WHEEL,sha256=6iYPr8vTHsyDK75jr9X0V3I9wPSVmtwr_8fdATBciGk,98
-ipex_llm-2.3.0b20250414.dist-info/entry_points.txt,sha256=TiUyBB2MRmfF3ko-pyAEzqeBCRnyhu27bNOAsWPp3e8,61
-ipex_llm-2.3.0b20250414.dist-info/top_level.txt,sha256=CGCMHM-SyqUabU4h8RqJ2KTYckQUO3LvIWwmUQ6Qbzw,9
-ipex_llm-2.3.0b20250414.dist-info/RECORD,,
+ipex_llm-2.3.0b20250417.data/scripts/ipex-llm-init.bat,sha256=HPtCYuDYwEatq7dAwOvdfVcHYCpAVdbj75K1qh0vQek,2578
+ipex_llm-2.3.0b20250417.data/scripts/llm-chat.ps1,sha256=6qrs-hGVAV8IKh7Jx8nq_XrnZcjd7qGU5wndArM7Yag,2769
+ipex_llm-2.3.0b20250417.data/scripts/llm-cli.ps1,sha256=3qBtTLs_EjYDnM8YyCpJhzLnGCKTEGssu9UNqfkjVXs,3009
+ipex_llm-2.3.0b20250417.dist-info/METADATA,sha256=kQC4s8NXPljYj5P64ZSaphyblBk1e2OuBJJeKStcPmo,13917
+ipex_llm-2.3.0b20250417.dist-info/WHEEL,sha256=6iYPr8vTHsyDK75jr9X0V3I9wPSVmtwr_8fdATBciGk,98
+ipex_llm-2.3.0b20250417.dist-info/entry_points.txt,sha256=TiUyBB2MRmfF3ko-pyAEzqeBCRnyhu27bNOAsWPp3e8,61
+ipex_llm-2.3.0b20250417.dist-info/top_level.txt,sha256=CGCMHM-SyqUabU4h8RqJ2KTYckQUO3LvIWwmUQ6Qbzw,9
+ipex_llm-2.3.0b20250417.dist-info/RECORD,,