Upload folder using huggingface_hub

config.json

@@ -1,4 +1,5 @@
 {
+  "_attn_implementation_autoset": true,
   "_name_or_path": "distil-whisper/distil-large-v3",
   "activation_dropout": 0.0,
   "activation_function": "gelu",
@@ -39,7 +40,7 @@
   "num_mel_bins": 128,
   "pad_token_id": 50256,
   "scale_embedding": false,
-  "transformers_version": "4.45.2",
+  "transformers_version": "4.46.3",
   "use_cache": true,
   "use_weighted_layer_sum": false,
   "vocab_size": 51866

generation_config.json

@@ -303,5 +303,5 @@
     "transcribe": 50360,
     "translate": 50359
   },
-  "transformers_version": "4.45.2"
+  "transformers_version": "4.46.3"
 }

openvino_decoder_model.xml

@@ -10,7 +10,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="0" name="Parameter_25934" type="Parameter" version="opset1">
+		<layer id="0" name="Parameter_26207" type="Parameter" version="opset1">
 			<data shape="?,?,?" element_type="f32" />
 			<output>
 				<port id="0" precision="FP32" names="encoder_hidden_states">
@@ -29,7 +29,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="3" name="Convert_42265" type="Convert" version="opset1">
+		<layer id="3" name="Convert_42605" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -53,7 +53,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="5" name="Convert_42268" type="Convert" version="opset1">
+		<layer id="5" name="Convert_42608" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -130,7 +130,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="10" name="Constant_31368" type="Const" version="opset1">
+		<layer id="10" name="Constant_31708" type="Const" version="opset1">
 			<data element_type="i64" shape="2" offset="66544078" size="16" />
 			<output>
 				<port id="0" precision="I64">
@@ -207,7 +207,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="16" name="Convert_42496" type="Convert" version="opset1">
+		<layer id="16" name="Convert_42836" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -231,7 +231,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="18" name="Convert_42499" type="Convert" version="opset1">
+		<layer id="18" name="Convert_42839" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -314,7 +314,7 @@
 				<port id="0" precision="I64" names="23" />
 			</output>
 		</layer>
-		<layer id="24" name="ShapeOf_31341" type="ShapeOf" version="opset3">
+		<layer id="24" name="ShapeOf_31624" type="ShapeOf" version="opset3">
 			<data output_type="i64" />
 			<input>
 				<port id="0" precision="I64">
@@ -328,19 +328,19 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="25" name="Constant_31342" type="Const" version="opset1">
+		<layer id="25" name="Constant_31625" type="Const" version="opset1">
 			<data element_type="i64" shape="" offset="67118890" size="8" />
 			<output>
 				<port id="0" precision="I64" />
 			</output>
 		</layer>
-		<layer id="26" name="Constant_31343" type="Const" version="opset1">
+		<layer id="26" name="Constant_31626" type="Const" version="opset1">
 			<data element_type="i64" shape="" offset="67118882" size="8" />
 			<output>
 				<port id="0" precision="I64" />
 			</output>
 		</layer>
-		<layer id="27" name="Gather_31344" type="Gather" version="opset8">
+		<layer id="27" name="Gather_31627" type="Gather" version="opset8">
 			<data batch_dims="0" />
 			<input>
 				<port id="0" precision="I64">
@@ -401,7 +401,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="32" name="Constant_25931" type="Const" version="opset1">
+		<layer id="32" name="__module.model.model.decoder.embed_positions/aten::index/Constant" type="Const" version="opset1">
 			<data element_type="i32" shape="" offset="66544094" size="4" />
 			<output>
 				<port id="0" precision="I32" />
@@ -478,7 +478,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="37" name="Constant_31109" type="Const" version="opset1">
+		<layer id="37" name="Constant_31390" type="Const" version="opset1">
 			<data element_type="f32" shape="1, 1, 1280" offset="67118906" size="5120" />
 			<output>
 				<port id="0" precision="FP32">
@@ -510,7 +510,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="39" name="Constant_31110" type="Const" version="opset1">
+		<layer id="39" name="Constant_31391" type="Const" version="opset1">
 			<data element_type="f32" shape="1, 1, 1280" offset="67124026" size="5120" />
 			<output>
 				<port id="0" precision="FP32">
@@ -542,7 +542,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="41" name="Constant_31012" type="Const" version="opset1">
+		<layer id="41" name="Constant_31293" type="Const" version="opset1">
 			<data element_type="u8" shape="1280, 1280" offset="67129146" size="1638400" />
 			<output>
 				<port id="0" precision="U8">
@@ -551,7 +551,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="42" name="Convert_42386" type="Convert" version="opset1">
+		<layer id="42" name="Convert_42726" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -566,7 +566,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="43" name="Constant_31012/zero_point" type="Const" version="opset1">
+		<layer id="43" name="Constant_31293/zero_point" type="Const" version="opset1">
 			<data element_type="u8" shape="1280, 1" offset="68767546" size="1280" />
 			<output>
 				<port id="0" precision="U8">
@@ -575,7 +575,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="44" name="Convert_42389" type="Convert" version="opset1">
+		<layer id="44" name="Convert_42729" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -590,7 +590,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="45" name="Constant_31012/zero_point/subtract" type="Subtract" version="opset1">
+		<layer id="45" name="Constant_31293/zero_point/subtract" type="Subtract" version="opset1">
 			<data auto_broadcast="numpy" />
 			<input>
 				<port id="0" precision="FP16">
@@ -609,7 +609,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="46" name="Constant_31012/scale" type="Const" version="opset1">
+		<layer id="46" name="Constant_31293/scale" type="Const" version="opset1">
 			<data element_type="f16" shape="1280, 1" offset="68768826" size="2560" />
 			<output>
 				<port id="0" precision="FP16">
@@ -618,7 +618,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="47" name="Constant_31012/fq_weights_1" type="Multiply" version="opset1">
+		<layer id="47" name="Constant_31293/fq_weights_1" type="Multiply" version="opset1">
 			<data auto_broadcast="numpy" />
 			<input>
 				<port id="0" precision="FP16">
@@ -637,7 +637,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="48" name="Constant_31012/fq_weights_1/convert" type="Convert" version="opset1">
+		<layer id="48" name="Constant_31293/fq_weights_1/convert" type="Convert" version="opset1">
 			<data destination_type="f32" />
 			<input>
 				<port id="0" precision="FP16">
@@ -652,7 +652,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="49" name="Multiply_30995" type="MatMul" version="opset1">
+		<layer id="49" name="Multiply_31276" type="MatMul" version="opset1">
 			<data transpose_a="false" transpose_b="true" />
 			<input>
 				<port id="0" precision="FP32">
@@ -673,7 +673,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="50" name="Constant_31111" type="Const" version="opset1">
+		<layer id="50" name="Constant_31392" type="Const" version="opset1">
 			<data element_type="f32" shape="1, 1, 1280" offset="68771386" size="5120" />
 			<output>
 				<port id="0" precision="FP32">
@@ -705,7 +705,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="52" name="Constant_31369" type="Const" version="opset1">
+		<layer id="52" name="Constant_31709" type="Const" version="opset1">
 			<data element_type="i64" shape="4" offset="68776506" size="32" />
 			<output>
 				<port id="0" precision="I64">
@@ -772,7 +772,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="57" name="Convert_42375" type="Convert" version="opset1">
+		<layer id="57" name="Convert_42715" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -796,7 +796,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="59" name="Convert_42378" type="Convert" version="opset1">
+		<layer id="59" name="Convert_42718" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -894,7 +894,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="65" name="Constant_31370" type="Const" version="opset1">
+		<layer id="65" name="Constant_31710" type="Const" version="opset1">
 			<data element_type="i64" shape="4" offset="68776506" size="32" />
 			<output>
 				<port id="0" precision="I64">
@@ -983,13 +983,13 @@
 				<port id="0" precision="I32" />
 			</output>
 		</layer>
-		<layer id="71" name="__module.model.model.decoder/aten::full/Convert" type="Const" version="opset1">
+		<layer id="71" name="17" type="Const" version="opset1">
 			<data element_type="f32" shape="" offset="70418794" size="4" />
 			<output>
-				<port id="0" precision="FP32" />
+				<port id="0" precision="FP32" names="17" />
 			</output>
 		</layer>
-		<layer id="72" name="Constant_31351" type="Const" version="opset1">
+		<layer id="72" name="Constant_31648" type="Const" version="opset1">
 			<data element_type="i64" shape="1" offset="67118890" size="8" />
 			<output>
 				<port id="0" precision="I64">
@@ -997,7 +997,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="73" name="Reshape_31352" type="Reshape" version="opset1">
+		<layer id="73" name="Reshape_31649" type="Reshape" version="opset1">
 			<data special_zero="false" />
 			<input>
 				<port id="0" precision="I64" />
@@ -1027,13 +1027,13 @@
 				<port id="2" precision="I64" names="62,64" />
 			</output>
 		</layer>
-		<layer id="76" name="Constant_23811" type="Const" version="opset1">
+		<layer id="76" name="Constant_24084" type="Const" version="opset1">
 			<data element_type="i32" shape="" offset="66544094" size="4" />
 			<output>
 				<port id="0" precision="I32" />
 			</output>
 		</layer>
-		<layer id="77" name="Unsqueeze_23812" type="Unsqueeze" version="opset1">
+		<layer id="77" name="Unsqueeze_24085" type="Unsqueeze" version="opset1">
 			<input>
 				<port id="0" precision="I64" />
 				<port id="1" precision="I32" />
@@ -1075,7 +1075,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="80" name="ShapeOf_31353" type="ShapeOf" version="opset3">
+		<layer id="80" name="ShapeOf_31655" type="ShapeOf" version="opset3">
 			<data output_type="i32" />
 			<input>
 				<port id="0" precision="FP32">
@@ -1089,19 +1089,19 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="81" name="Constant_31354" type="Const" version="opset1">
+		<layer id="81" name="Constant_31656" type="Const" version="opset1">
 			<data element_type="i64" shape="" offset="67118890" size="8" />
 			<output>
 				<port id="0" precision="I64" />
 			</output>
 		</layer>
-		<layer id="82" name="Constant_31355" type="Const" version="opset1">
+		<layer id="82" name="Constant_31657" type="Const" version="opset1">
 			<data element_type="i64" shape="" offset="67118882" size="8" />
 			<output>
 				<port id="0" precision="I64" />
 			</output>
 		</layer>
-		<layer id="83" name="Gather_31356" type="Gather" version="opset8">
+		<layer id="83" name="Gather_31658" type="Gather" version="opset8">
 			<data batch_dims="0" />
 			<input>
 				<port id="0" precision="I32">
@@ -1153,7 +1153,7 @@
 				<port id="0" precision="I32" />
 			</output>
 		</layer>
-		<layer id="88" name="Convert_31359" type="Convert" version="opset1">
+		<layer id="88" name="Convert_31661" type="Convert" version="opset1">
 			<data destination_type="i32" />
 			<input>
 				<port id="0" precision="I64" />
@@ -1295,7 +1295,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="100" name="Constant_23914" type="Const" version="opset1">
+		<layer id="100" name="Constant_24187" type="Const" version="opset1">
 			<data element_type="i64" shape="2" offset="70418798" size="16" />
 			<output>
 				<port id="0" precision="I64" names="70">
@@ -1412,7 +1412,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="108" name="Constant_31348" type="Const" version="opset1">
+		<layer id="108" name="Constant_31639" type="Const" version="opset1">
 			<data element_type="i64" shape="1" offset="67118882" size="8" />
 			<output>
 				<port id="0" precision="I64">
@@ -1420,13 +1420,13 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="109" name="Constant_31349" type="Const" version="opset1">
+		<layer id="109" name="Constant_31640" type="Const" version="opset1">
 			<data element_type="i64" shape="" offset="67118882" size="8" />
 			<output>
 				<port id="0" precision="I64" />
 			</output>
 		</layer>
-		<layer id="110" name="Gather_31350" type="Gather" version="opset8">
+		<layer id="110" name="Gather_31641" type="Gather" version="opset8">
 			<data batch_dims="0" />
 			<input>
 				<port id="0" precision="I64">
@@ -1443,7 +1443,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="111" name="Constant_30340" type="Const" version="opset1">
+		<layer id="111" name="Constant_30625" type="Const" version="opset1">
 			<data element_type="i64" shape="1" offset="67118890" size="8" />
 			<output>
 				<port id="0" precision="I64">
@@ -1451,7 +1451,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="112" name="Constant_30342" type="Const" version="opset1">
+		<layer id="112" name="Constant_30627" type="Const" version="opset1">
 			<data element_type="i64" shape="1" offset="67118890" size="8" />
 			<output>
 				<port id="0" precision="I64">
@@ -1459,7 +1459,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="113" name="Constant_30344" type="Const" version="opset1">
+		<layer id="113" name="Constant_30629" type="Const" version="opset1">
 			<data element_type="i64" shape="1" offset="67118890" size="8" />
 			<output>
 				<port id="0" precision="I64">
@@ -1511,7 +1511,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="116" name="Constant_24348" type="Const" version="opset1">
+		<layer id="116" name="Constant_24621" type="Const" version="opset1">
 			<data element_type="i64" shape="1" offset="67118882" size="8" />
 			<output>
 				<port id="0" precision="I64">
@@ -1519,7 +1519,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="117" name="ShapeOf_31364" type="ShapeOf" version="opset3">
+		<layer id="117" name="ShapeOf_31670" type="ShapeOf" version="opset3">
 			<data output_type="i64" />
 			<input>
 				<port id="0" precision="FP32">
@@ -1534,7 +1534,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="118" name="Constant_31365" type="Const" version="opset1">
+		<layer id="118" name="Constant_31671" type="Const" version="opset1">
 			<data element_type="i64" shape="1" offset="67118890" size="8" />
 			<output>
 				<port id="0" precision="I64">
@@ -1542,13 +1542,13 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="119" name="Constant_31366" type="Const" version="opset1">
+		<layer id="119" name="Constant_31672" type="Const" version="opset1">
 			<data element_type="i64" shape="" offset="67118882" size="8" />
 			<output>
 				<port id="0" precision="I64" />
 			</output>
 		</layer>
-		<layer id="120" name="Gather_31367" type="Gather" version="opset8">
+		<layer id="120" name="Gather_31673" type="Gather" version="opset8">
 			<data batch_dims="0" />
 			<input>
 				<port id="0" precision="I64">
@@ -1565,7 +1565,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="121" name="Constant_24351" type="Const" version="opset1">
+		<layer id="121" name="Constant_24624" type="Const" version="opset1">
 			<data element_type="i64" shape="1" offset="67118890" size="8" />
 			<output>
 				<port id="0" precision="I64">
@@ -1664,7 +1664,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="127" name="Convert_42364" type="Convert" version="opset1">
+		<layer id="127" name="Convert_42704" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -1688,7 +1688,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="129" name="Convert_42367" type="Convert" version="opset1">
+		<layer id="129" name="Convert_42707" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -1786,7 +1786,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="135" name="Constant_31112" type="Const" version="opset1">
+		<layer id="135" name="Constant_31393" type="Const" version="opset1">
 			<data element_type="f32" shape="1, 1, 1280" offset="72061062" size="5120" />
 			<output>
 				<port id="0" precision="FP32">
@@ -1818,7 +1818,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="137" name="Constant_31371" type="Const" version="opset1">
+		<layer id="137" name="Constant_31711" type="Const" version="opset1">
 			<data element_type="i64" shape="4" offset="68776506" size="32" />
 			<output>
 				<port id="0" precision="I64">
@@ -1930,7 +1930,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="144" name="Constant_31372" type="Const" version="opset1">
+		<layer id="144" name="Constant_31712" type="Const" version="opset1">
 			<data element_type="i64" shape="3" offset="72066182" size="24" />
 			<output>
 				<port id="0" precision="I64">
@@ -1968,7 +1968,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="147" name="Convert_42397" type="Convert" version="opset1">
+		<layer id="147" name="Convert_42737" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -1992,7 +1992,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="149" name="Convert_42400" type="Convert" version="opset1">
+		<layer id="149" name="Convert_42740" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -2090,7 +2090,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="155" name="Constant_31113" type="Const" version="opset1">
+		<layer id="155" name="Constant_31394" type="Const" version="opset1">
 			<data element_type="f32" shape="1, 1, 1280" offset="73708446" size="5120" />
 			<output>
 				<port id="0" precision="FP32">
@@ -2172,7 +2172,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="160" name="Constant_31114" type="Const" version="opset1">
+		<layer id="160" name="Constant_31395" type="Const" version="opset1">
 			<data element_type="f32" shape="1, 1, 1280" offset="73713566" size="5120" />
 			<output>
 				<port id="0" precision="FP32">
@@ -2204,7 +2204,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="162" name="Constant_31115" type="Const" version="opset1">
+		<layer id="162" name="Constant_31396" type="Const" version="opset1">
 			<data element_type="f32" shape="1, 1, 1280" offset="73718686" size="5120" />
 			<output>
 				<port id="0" precision="FP32">
@@ -2236,7 +2236,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="164" name="Constant_31015" type="Const" version="opset1">
+		<layer id="164" name="Constant_31296" type="Const" version="opset1">
 			<data element_type="u8" shape="1280, 1280" offset="73723806" size="1638400" />
 			<output>
 				<port id="0" precision="U8">
@@ -2245,7 +2245,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="165" name="Convert_42408" type="Convert" version="opset1">
+		<layer id="165" name="Convert_42748" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -2260,7 +2260,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="166" name="Constant_31015/zero_point" type="Const" version="opset1">
+		<layer id="166" name="Constant_31296/zero_point" type="Const" version="opset1">
 			<data element_type="u8" shape="1280, 1" offset="75362206" size="1280" />
 			<output>
 				<port id="0" precision="U8">
@@ -2269,7 +2269,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="167" name="Convert_42411" type="Convert" version="opset1">
+		<layer id="167" name="Convert_42751" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -2284,7 +2284,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="168" name="Constant_31015/zero_point/subtract" type="Subtract" version="opset1">
+		<layer id="168" name="Constant_31296/zero_point/subtract" type="Subtract" version="opset1">
 			<data auto_broadcast="numpy" />
 			<input>
 				<port id="0" precision="FP16">
@@ -2303,7 +2303,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="169" name="Constant_31015/scale" type="Const" version="opset1">
+		<layer id="169" name="Constant_31296/scale" type="Const" version="opset1">
 			<data element_type="f16" shape="1280, 1" offset="75363486" size="2560" />
 			<output>
 				<port id="0" precision="FP16">
@@ -2312,7 +2312,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="170" name="Constant_31015/fq_weights_1" type="Multiply" version="opset1">
+		<layer id="170" name="Constant_31296/fq_weights_1" type="Multiply" version="opset1">
 			<data auto_broadcast="numpy" />
 			<input>
 				<port id="0" precision="FP16">
@@ -2331,7 +2331,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="171" name="Constant_31015/fq_weights_1/convert" type="Convert" version="opset1">
+		<layer id="171" name="Constant_31296/fq_weights_1/convert" type="Convert" version="opset1">
 			<data destination_type="f32" />
 			<input>
 				<port id="0" precision="FP16">
@@ -2346,7 +2346,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="172" name="Multiply_30999" type="MatMul" version="opset1">
+		<layer id="172" name="Multiply_31280" type="MatMul" version="opset1">
 			<data transpose_a="false" transpose_b="true" />
 			<input>
 				<port id="0" precision="FP32">
@@ -2367,7 +2367,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="173" name="Constant_31116" type="Const" version="opset1">
+		<layer id="173" name="Constant_31397" type="Const" version="opset1">
 			<data element_type="f32" shape="1, 1, 1280" offset="75366046" size="5120" />
 			<output>
 				<port id="0" precision="FP32">
@@ -2399,7 +2399,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="175" name="Constant_31373" type="Const" version="opset1">
+		<layer id="175" name="Constant_31713" type="Const" version="opset1">
 			<data element_type="i64" shape="4" offset="68776506" size="32" />
 			<output>
 				<port id="0" precision="I64">
@@ -2466,7 +2466,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="180" name="Convert_42353" type="Convert" version="opset1">
+		<layer id="180" name="Convert_42693" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -2490,7 +2490,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="182" name="Convert_42356" type="Convert" version="opset1">
+		<layer id="182" name="Convert_42696" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -2588,7 +2588,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="188" name="Constant_24572" type="Const" version="opset1">
+		<layer id="188" name="Constant_24369" type="Const" version="opset1">
 			<data element_type="i64" shape="1" offset="77013406" size="8" />
 			<output>
 				<port id="0" precision="I64">
@@ -2596,7 +2596,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="189" name="Constant_24573" type="Const" version="opset1">
+		<layer id="189" name="Constant_24370" type="Const" version="opset1">
 			<data element_type="i64" shape="1" offset="77013414" size="8" />
 			<output>
 				<port id="0" precision="I64">
@@ -2604,7 +2604,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="190" name="Constant_24574" type="Const" version="opset1">
+		<layer id="190" name="Constant_24371" type="Const" version="opset1">
 			<data element_type="i64" shape="1" offset="77013422" size="8" />
 			<output>
 				<port id="0" precision="I64">
@@ -2612,7 +2612,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="191" name="__module.model.model.decoder.layers.0.encoder_attn/prim::ListConstruct_1" type="Concat" version="opset1">
+		<layer id="191" name="__module.model.model.decoder.layers.0.self_attn/prim::ListConstruct_1" type="Concat" version="opset1">
 			<data axis="0" />
 			<input>
 				<port id="0" precision="I64">
@@ -2737,7 +2737,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="198" name="Convert_42342" type="Convert" version="opset1">
+		<layer id="198" name="Convert_42682" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -2761,7 +2761,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="200" name="Convert_42345" type="Convert" version="opset1">
+		<layer id="200" name="Convert_42685" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -2859,7 +2859,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="206" name="Constant_31117" type="Const" version="opset1">
+		<layer id="206" name="Constant_31398" type="Const" version="opset1">
 			<data element_type="f32" shape="1, 1, 1280" offset="78655670" size="5120" />
 			<output>
 				<port id="0" precision="FP32">
@@ -2995,7 +2995,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="214" name="Constant_31374" type="Const" version="opset1">
+		<layer id="214" name="Constant_31714" type="Const" version="opset1">
 			<data element_type="i64" shape="3" offset="78660790" size="24" />
 			<output>
 				<port id="0" precision="I64">
@@ -3033,7 +3033,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="217" name="Convert_42419" type="Convert" version="opset1">
+		<layer id="217" name="Convert_42759" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -3057,7 +3057,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="219" name="Convert_42422" type="Convert" version="opset1">
+		<layer id="219" name="Convert_42762" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -3155,7 +3155,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="225" name="Constant_31118" type="Const" version="opset1">
+		<layer id="225" name="Constant_31399" type="Const" version="opset1">
 			<data element_type="f32" shape="1, 1, 1280" offset="80303054" size="5120" />
 			<output>
 				<port id="0" precision="FP32">
@@ -3237,7 +3237,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="230" name="Constant_31119" type="Const" version="opset1">
+		<layer id="230" name="Constant_31400" type="Const" version="opset1">
 			<data element_type="f32" shape="1, 1, 1280" offset="80308174" size="5120" />
 			<output>
 				<port id="0" precision="FP32">
@@ -3269,7 +3269,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="232" name="Constant_31120" type="Const" version="opset1">
+		<layer id="232" name="Constant_31401" type="Const" version="opset1">
 			<data element_type="f32" shape="1, 1, 1280" offset="80313294" size="5120" />
 			<output>
 				<port id="0" precision="FP32">
@@ -3310,7 +3310,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="235" name="Convert_42276" type="Convert" version="opset1">
+		<layer id="235" name="Convert_42616" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -3334,7 +3334,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="237" name="Convert_42279" type="Convert" version="opset1">
+		<layer id="237" name="Convert_42619" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -3432,7 +3432,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="243" name="Constant_31121" type="Const" version="opset1">
+		<layer id="243" name="Constant_31402" type="Const" version="opset1">
 			<data element_type="f32" shape="1, 1, 5120" offset="86887374" size="20480" />
 			<output>
 				<port id="0" precision="FP32">
@@ -3490,7 +3490,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="247" name="Convert_42287" type="Convert" version="opset1">
+		<layer id="247" name="Convert_42627" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -3514,7 +3514,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="249" name="Convert_42290" type="Convert" version="opset1">
+		<layer id="249" name="Convert_42630" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -3612,7 +3612,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="255" name="Constant_31122" type="Const" version="opset1">
+		<layer id="255" name="Constant_31403" type="Const" version="opset1">
 			<data element_type="f32" shape="1, 1, 1280" offset="93465294" size="5120" />
 			<output>
 				<port id="0" precision="FP32">
@@ -3694,7 +3694,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="260" name="Constant_31123" type="Const" version="opset1">
+		<layer id="260" name="Constant_31404" type="Const" version="opset1">
 			<data element_type="f32" shape="1, 1, 1280" offset="93470414" size="5120" />
 			<output>
 				<port id="0" precision="FP32">
@@ -3726,7 +3726,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="262" name="Constant_31124" type="Const" version="opset1">
+		<layer id="262" name="Constant_31405" type="Const" version="opset1">
 			<data element_type="f32" shape="1, 1, 1280" offset="93475534" size="5120" />
 			<output>
 				<port id="0" precision="FP32">
@@ -3758,7 +3758,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="264" name="Constant_31018" type="Const" version="opset1">
+		<layer id="264" name="Constant_31299" type="Const" version="opset1">
 			<data element_type="u8" shape="1280, 1280" offset="93480654" size="1638400" />
 			<output>
 				<port id="0" precision="U8">
@@ -3767,7 +3767,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="265" name="Convert_42452" type="Convert" version="opset1">
+		<layer id="265" name="Convert_42792" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -3782,7 +3782,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="266" name="Constant_31018/zero_point" type="Const" version="opset1">
+		<layer id="266" name="Constant_31299/zero_point" type="Const" version="opset1">
 			<data element_type="u8" shape="1280, 1" offset="95119054" size="1280" />
 			<output>
 				<port id="0" precision="U8">
@@ -3791,7 +3791,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="267" name="Convert_42455" type="Convert" version="opset1">
+		<layer id="267" name="Convert_42795" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -3806,7 +3806,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="268" name="Constant_31018/zero_point/subtract" type="Subtract" version="opset1">
+		<layer id="268" name="Constant_31299/zero_point/subtract" type="Subtract" version="opset1">
 			<data auto_broadcast="numpy" />
 			<input>
 				<port id="0" precision="FP16">
@@ -3825,7 +3825,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="269" name="Constant_31018/scale" type="Const" version="opset1">
+		<layer id="269" name="Constant_31299/scale" type="Const" version="opset1">
 			<data element_type="f16" shape="1280, 1" offset="95120334" size="2560" />
 			<output>
 				<port id="0" precision="FP16">
@@ -3834,7 +3834,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="270" name="Constant_31018/fq_weights_1" type="Multiply" version="opset1">
+		<layer id="270" name="Constant_31299/fq_weights_1" type="Multiply" version="opset1">
 			<data auto_broadcast="numpy" />
 			<input>
 				<port id="0" precision="FP16">
@@ -3853,7 +3853,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="271" name="Constant_31018/fq_weights_1/convert" type="Convert" version="opset1">
+		<layer id="271" name="Constant_31299/fq_weights_1/convert" type="Convert" version="opset1">
 			<data destination_type="f32" />
 			<input>
 				<port id="0" precision="FP16">
@@ -3868,7 +3868,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="272" name="Multiply_31003" type="MatMul" version="opset1">
+		<layer id="272" name="Multiply_31284" type="MatMul" version="opset1">
 			<data transpose_a="false" transpose_b="true" />
 			<input>
 				<port id="0" precision="FP32">
@@ -3889,7 +3889,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="273" name="Constant_31125" type="Const" version="opset1">
+		<layer id="273" name="Constant_31406" type="Const" version="opset1">
 			<data element_type="f32" shape="1, 1, 1280" offset="95122894" size="5120" />
 			<output>
 				<port id="0" precision="FP32">
@@ -3921,7 +3921,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="275" name="Constant_31375" type="Const" version="opset1">
+		<layer id="275" name="Constant_31715" type="Const" version="opset1">
 			<data element_type="i64" shape="4" offset="68776506" size="32" />
 			<output>
 				<port id="0" precision="I64">
@@ -3988,7 +3988,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="280" name="Convert_42441" type="Convert" version="opset1">
+		<layer id="280" name="Convert_42781" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -4012,7 +4012,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="282" name="Convert_42444" type="Convert" version="opset1">
+		<layer id="282" name="Convert_42784" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -4110,7 +4110,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="288" name="Constant_31376" type="Const" version="opset1">
+		<layer id="288" name="Constant_31716" type="Const" version="opset1">
 			<data element_type="i64" shape="4" offset="68776506" size="32" />
 			<output>
 				<port id="0" precision="I64">
@@ -4246,7 +4246,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="296" name="Convert_42430" type="Convert" version="opset1">
+		<layer id="296" name="Convert_42770" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -4270,7 +4270,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="298" name="Convert_42433" type="Convert" version="opset1">
+		<layer id="298" name="Convert_42773" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -4368,7 +4368,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="304" name="Constant_31126" type="Const" version="opset1">
+		<layer id="304" name="Constant_31407" type="Const" version="opset1">
 			<data element_type="f32" shape="1, 1, 1280" offset="98412494" size="5120" />
 			<output>
 				<port id="0" precision="FP32">
@@ -4400,7 +4400,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="306" name="Constant_31377" type="Const" version="opset1">
+		<layer id="306" name="Constant_31717" type="Const" version="opset1">
 			<data element_type="i64" shape="4" offset="68776506" size="32" />
 			<output>
 				<port id="0" precision="I64">
@@ -4512,7 +4512,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="313" name="Constant_31378" type="Const" version="opset1">
+		<layer id="313" name="Constant_31718" type="Const" version="opset1">
 			<data element_type="i64" shape="3" offset="72066182" size="24" />
 			<output>
 				<port id="0" precision="I64">
@@ -4550,7 +4550,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="316" name="Convert_42463" type="Convert" version="opset1">
+		<layer id="316" name="Convert_42803" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -4574,7 +4574,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="318" name="Convert_42466" type="Convert" version="opset1">
+		<layer id="318" name="Convert_42806" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -4672,7 +4672,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="324" name="Constant_31127" type="Const" version="opset1">
+		<layer id="324" name="Constant_31408" type="Const" version="opset1">
 			<data element_type="f32" shape="1, 1, 1280" offset="100059854" size="5120" />
 			<output>
 				<port id="0" precision="FP32">
@@ -4754,7 +4754,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="329" name="Constant_31128" type="Const" version="opset1">
+		<layer id="329" name="Constant_31409" type="Const" version="opset1">
 			<data element_type="f32" shape="1, 1, 1280" offset="100064974" size="5120" />
 			<output>
 				<port id="0" precision="FP32">
@@ -4786,7 +4786,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="331" name="Constant_31129" type="Const" version="opset1">
+		<layer id="331" name="Constant_31410" type="Const" version="opset1">
 			<data element_type="f32" shape="1, 1, 1280" offset="100070094" size="5120" />
 			<output>
 				<port id="0" precision="FP32">
@@ -4818,7 +4818,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="333" name="Constant_31021" type="Const" version="opset1">
+		<layer id="333" name="Constant_31302" type="Const" version="opset1">
 			<data element_type="u8" shape="1280, 1280" offset="100075214" size="1638400" />
 			<output>
 				<port id="0" precision="U8">
@@ -4827,7 +4827,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="334" name="Convert_42474" type="Convert" version="opset1">
+		<layer id="334" name="Convert_42814" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -4842,7 +4842,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="335" name="Constant_31021/zero_point" type="Const" version="opset1">
+		<layer id="335" name="Constant_31302/zero_point" type="Const" version="opset1">
 			<data element_type="u8" shape="1280, 1" offset="101713614" size="1280" />
 			<output>
 				<port id="0" precision="U8">
@@ -4851,7 +4851,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="336" name="Convert_42477" type="Convert" version="opset1">
+		<layer id="336" name="Convert_42817" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -4866,7 +4866,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="337" name="Constant_31021/zero_point/subtract" type="Subtract" version="opset1">
+		<layer id="337" name="Constant_31302/zero_point/subtract" type="Subtract" version="opset1">
 			<data auto_broadcast="numpy" />
 			<input>
 				<port id="0" precision="FP16">
@@ -4885,7 +4885,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="338" name="Constant_31021/scale" type="Const" version="opset1">
+		<layer id="338" name="Constant_31302/scale" type="Const" version="opset1">
 			<data element_type="f16" shape="1280, 1" offset="101714894" size="2560" />
 			<output>
 				<port id="0" precision="FP16">
@@ -4894,7 +4894,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="339" name="Constant_31021/fq_weights_1" type="Multiply" version="opset1">
+		<layer id="339" name="Constant_31302/fq_weights_1" type="Multiply" version="opset1">
 			<data auto_broadcast="numpy" />
 			<input>
 				<port id="0" precision="FP16">
@@ -4913,7 +4913,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="340" name="Constant_31021/fq_weights_1/convert" type="Convert" version="opset1">
+		<layer id="340" name="Constant_31302/fq_weights_1/convert" type="Convert" version="opset1">
 			<data destination_type="f32" />
 			<input>
 				<port id="0" precision="FP16">
@@ -4928,7 +4928,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="341" name="Multiply_31007" type="MatMul" version="opset1">
+		<layer id="341" name="Multiply_31288" type="MatMul" version="opset1">
 			<data transpose_a="false" transpose_b="true" />
 			<input>
 				<port id="0" precision="FP32">
@@ -4949,7 +4949,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="342" name="Constant_31130" type="Const" version="opset1">
+		<layer id="342" name="Constant_31411" type="Const" version="opset1">
 			<data element_type="f32" shape="1, 1, 1280" offset="101717454" size="5120" />
 			<output>
 				<port id="0" precision="FP32">
@@ -4981,7 +4981,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="344" name="Constant_31379" type="Const" version="opset1">
+		<layer id="344" name="Constant_31719" type="Const" version="opset1">
 			<data element_type="i64" shape="4" offset="68776506" size="32" />
 			<output>
 				<port id="0" precision="I64">
@@ -5048,7 +5048,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="349" name="Convert_42331" type="Convert" version="opset1">
+		<layer id="349" name="Convert_42671" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -5072,7 +5072,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="351" name="Convert_42334" type="Convert" version="opset1">
+		<layer id="351" name="Convert_42674" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -5273,7 +5273,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="363" name="Convert_42320" type="Convert" version="opset1">
+		<layer id="363" name="Convert_42660" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -5297,7 +5297,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="365" name="Convert_42323" type="Convert" version="opset1">
+		<layer id="365" name="Convert_42663" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -5395,7 +5395,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="371" name="Constant_31131" type="Const" version="opset1">
+		<layer id="371" name="Constant_31412" type="Const" version="opset1">
 			<data element_type="f32" shape="1, 1, 1280" offset="105007054" size="5120" />
 			<output>
 				<port id="0" precision="FP32">
@@ -5531,7 +5531,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="379" name="Constant_31380" type="Const" version="opset1">
+		<layer id="379" name="Constant_31720" type="Const" version="opset1">
 			<data element_type="i64" shape="3" offset="78660790" size="24" />
 			<output>
 				<port id="0" precision="I64">
@@ -5569,7 +5569,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="382" name="Convert_42485" type="Convert" version="opset1">
+		<layer id="382" name="Convert_42825" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -5593,7 +5593,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="384" name="Convert_42488" type="Convert" version="opset1">
+		<layer id="384" name="Convert_42828" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -5691,7 +5691,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="390" name="Constant_31132" type="Const" version="opset1">
+		<layer id="390" name="Constant_31413" type="Const" version="opset1">
 			<data element_type="f32" shape="1, 1, 1280" offset="106654414" size="5120" />
 			<output>
 				<port id="0" precision="FP32">
@@ -5773,7 +5773,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="395" name="Constant_31133" type="Const" version="opset1">
+		<layer id="395" name="Constant_31414" type="Const" version="opset1">
 			<data element_type="f32" shape="1, 1, 1280" offset="106659534" size="5120" />
 			<output>
 				<port id="0" precision="FP32">
@@ -5805,7 +5805,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="397" name="Constant_31134" type="Const" version="opset1">
+		<layer id="397" name="Constant_31415" type="Const" version="opset1">
 			<data element_type="f32" shape="1, 1, 1280" offset="106664654" size="5120" />
 			<output>
 				<port id="0" precision="FP32">
@@ -5846,7 +5846,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="400" name="Convert_42298" type="Convert" version="opset1">
+		<layer id="400" name="Convert_42638" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -5870,7 +5870,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="402" name="Convert_42301" type="Convert" version="opset1">
+		<layer id="402" name="Convert_42641" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -5968,7 +5968,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="408" name="Constant_31135" type="Const" version="opset1">
+		<layer id="408" name="Constant_31416" type="Const" version="opset1">
 			<data element_type="f32" shape="1, 1, 5120" offset="113238734" size="20480" />
 			<output>
 				<port id="0" precision="FP32">
@@ -6026,7 +6026,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="412" name="Convert_42309" type="Convert" version="opset1">
+		<layer id="412" name="Convert_42649" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -6050,7 +6050,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="414" name="Convert_42312" type="Convert" version="opset1">
+		<layer id="414" name="Convert_42652" type="Convert" version="opset1">
 			<data destination_type="f16" />
 			<input>
 				<port id="0" precision="U8">
@@ -6148,7 +6148,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="420" name="Constant_31136" type="Const" version="opset1">
+		<layer id="420" name="Constant_31417" type="Const" version="opset1">
 			<data element_type="f32" shape="1, 1, 1280" offset="119816654" size="5120" />
 			<output>
 				<port id="0" precision="FP32">
@@ -6230,7 +6230,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="425" name="Constant_31137" type="Const" version="opset1">
+		<layer id="425" name="Constant_31418" type="Const" version="opset1">
 			<data element_type="f32" shape="1, 1, 1280" offset="119821774" size="5120" />
 			<output>
 				<port id="0" precision="FP32">
@@ -6262,7 +6262,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="427" name="Constant_31138" type="Const" version="opset1">
+		<layer id="427" name="Constant_31419" type="Const" version="opset1">
 			<data element_type="f32" shape="1, 1, 1280" offset="119826894" size="5120" />
 			<output>
 				<port id="0" precision="FP32">
@@ -6315,7 +6315,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="430" name="Result_25933" type="Result" version="opset1">
+		<layer id="430" name="Result_26206" type="Result" version="opset1">
 			<input>
 				<port id="0" precision="FP32">
 					<dim>-1</dim>
@@ -6792,7 +6792,7 @@
 		<edge from-layer="429" from-port="2" to-layer="430" to-port="0" />
 	</edges>
 	<rt_info>
-		<Runtime_version value="2024.5.0-16901-32aaa2fbd96" />
+		<Runtime_version value="2024.5.0-17285-ea5c1dcfdf9-releases/2024/5" />
 		<conversion_parameters>
 			<framework value="pytorch" />
 			<is_python_object value="True" />
@@ -6800,9 +6800,10 @@
 		<nncf>
 			<friendly_names_were_updated value="True" />
 			<weight_compression>
-				<advanced_parameters value="{'awq_params': {'subset_size': 32, 'percent_to_apply': 0.002, 'alpha_min': 0.0, 'alpha_max': 1.0, 'steps': 100}, 'scale_estimation_params': {'subset_size': 64, 'initial_steps': 5, 'scale_steps': 5, 'weight_penalty': -1.0}, 'gptq_params': {'damp_percent': 0.1, 'block_size': 128, 'subset_size': 128}, 'lora_correction_params': {'adapter_rank': 8, 'num_iterations': 3, 'apply_regularization': True, 'subset_size': 128, 'use_int8_adapters': True}}" />
+				<advanced_parameters value="{'statistics_path': None, 'awq_params': {'subset_size': 32, 'percent_to_apply': 0.002, 'alpha_min': 0.0, 'alpha_max': 1.0, 'steps': 100}, 'scale_estimation_params': {'subset_size': 64, 'initial_steps': 5, 'scale_steps': 5, 'weight_penalty': -1.0}, 'gptq_params': {'damp_percent': 0.1, 'block_size': 128, 'subset_size': 128}, 'lora_correction_params': {'adapter_rank': 8, 'num_iterations': 3, 'apply_regularization': True, 'subset_size': 128, 'use_int8_adapters': True}}" />
 				<all_layers value="False" />
 				<awq value="False" />
+				<backup_mode value="int8_asym" />
 				<gptq value="False" />
 				<group_size value="-1" />
 				<ignored_scope value="[]" />
@@ -6814,10 +6815,10 @@
 			</weight_compression>
 		</nncf>
 		<optimum>
-			<optimum_intel_version value="1.20.0.dev0+2559620" />
-			<optimum_version value="1.23.1" />
-			<pytorch_version value="2.5.0" />
-			<transformers_version value="4.45.2" />
+			<optimum_intel_version value="1.21.0.dev0+d357376" />
+			<optimum_version value="1.23.3" />
+			<pytorch_version value="2.5.1" />
+			<transformers_version value="4.46.3" />
 		</optimum>
 	</rt_info>
 </net>

openvino_detokenizer.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b55ac04e90f19fede391281a9e5a90169fc646e1e302fe1208c782282b51ceaa
+oid sha256:1c95aa7a4b53ecb3768d2a6608aa9e419059aef1c75b9b99e6b1c56328281c18
 size 528306

openvino_detokenizer.xml

@@ -1,16 +1,16 @@
 <?xml version="1.0"?>
 <net name="detokenizer" version="11">
 	<layers>
-		<layer id="0" name="Parameter_32420" type="Parameter" version="opset1">
+		<layer id="0" name="Parameter_32760" type="Parameter" version="opset1">
 			<data shape="?,?" element_type="i64" />
 			<output>
-				<port id="0" precision="I64" names="Parameter_32420">
+				<port id="0" precision="I64" names="Parameter_32760">
 					<dim>-1</dim>
 					<dim>-1</dim>
 				</port>
 			</output>
 		</layer>
-		<layer id="1" name="Convert_32436" type="Convert" version="opset1">
+		<layer id="1" name="Convert_32776" type="Convert" version="opset1">
 			<data destination_type="i32" />
 			<input>
 				<port id="0" precision="I64">
@@ -25,7 +25,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="2" name="Constant_32381" type="Const" version="opset1">
+		<layer id="2" name="Constant_32721" type="Const" version="opset1">
 			<data element_type="u8" shape="528257" offset="0" size="528257" />
 			<output>
 				<port id="0" precision="U8">
@@ -33,7 +33,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="3" name="StringTensorUnpack_32382" type="StringTensorUnpack" version="extension">
+		<layer id="3" name="StringTensorUnpack_32722" type="StringTensorUnpack" version="extension">
 			<data mode="begins_ends" />
 			<input>
 				<port id="0" precision="U8">
@@ -52,7 +52,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="4" name="VocabDecoder_32421" type="VocabDecoder" version="extension">
+		<layer id="4" name="VocabDecoder_32761" type="VocabDecoder" version="extension">
 			<data skip_tokens="50257, 50258, 50259, 50260, 50261, 50262, 50263, 50264, 50265, 50266, 50267, 50268, 50269, 50270, 50271, 50272, 50273, 50274, 50275, 50276, 50277, 50278, 50279, 50280, 50281, 50282, 50283, 50284, 50285, 50286, 50287, 50288, 50289, 50290, 50291, 50292, 50293, 50294, 50295, 50296, 50297, 50298, 50299, 50300, 50301, 50302, 50303, 50304, 50305, 50306, 50307, 50308, 50309, 50310, 50311, 50312, 50313, 50314, 50315, 50316, 50317, 50318, 50319, 50320, 50321, 50322, 50323, 50324, 50325, 50326, 50327, 50328, 50329, 50330, 50331, 50332, 50333, 50334, 50335, 50336, 50337, 50338, 50339, 50340, 50341, 50342, 50343, 50344, 50345, 50346, 50347, 50348, 50349, 50350, 50351, 50352, 50353, 50354, 50355, 50356, 50357, 50358, 50359, 50360, 50361, 50362, 50363, 50364" />
 			<input>
 				<port id="0" precision="I32">
@@ -87,7 +87,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="5" name="FuzeRagged_32422" type="FuzeRagged" version="extension">
+		<layer id="5" name="FuzeRagged_32762" type="FuzeRagged" version="extension">
 			<input>
 				<port id="0" precision="I32">
 					<dim>-1</dim>
@@ -111,7 +111,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="6" name="Constant_32424" type="Const" version="opset1">
+		<layer id="6" name="Constant_32764" type="Const" version="opset1">
 			<data element_type="u8" shape="47" offset="528257" size="47" />
 			<output>
 				<port id="0" precision="U8">
@@ -119,7 +119,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="7" name="Constant_32426" type="Const" version="opset1">
+		<layer id="7" name="Constant_32766" type="Const" version="opset1">
 			<data element_type="u8" shape="2" offset="528304" size="2" />
 			<output>
 				<port id="0" precision="U8">
@@ -127,7 +127,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="8" name="RegexNormalization_32427" type="RegexNormalization" version="extension">
+		<layer id="8" name="RegexNormalization_32767" type="RegexNormalization" version="extension">
 			<data global_replace="true" />
 			<input>
 				<port id="0" precision="I32">
@@ -158,7 +158,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="9" name="StringTensorPack_32428" type="StringTensorPack" version="extension">
+		<layer id="9" name="StringTensorPack_32768" type="StringTensorPack" version="extension">
 			<data mode="begins_ends" />
 			<input>
 				<port id="0" precision="I32">
@@ -177,7 +177,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="10" name="Result_32429" type="Result" version="opset1">
+		<layer id="10" name="Result_32769" type="Result" version="opset1">
 			<input>
 				<port id="0" precision="STRING">
 					<dim>-1</dim>
@@ -207,9 +207,29 @@
 		<edge from-layer="9" from-port="3" to-layer="10" to-port="0" />
 	</edges>
 	<rt_info>
+		<add_attention_mask value="True" />
+		<add_prefix_space />
+		<add_special_tokens value="True" />
 		<bos_token_id value="50257" />
+		<clean_up_tokenization_spaces />
+		<detokenizer_input_type value="i64" />
 		<eos_token_id value="50257" />
+		<handle_special_tokens_with_re />
+		<number_of_inputs value="1" />
+		<openvino_tokenizers_version value="2024.5.0.0" />
+		<openvino_version value="2024.5.0" />
 		<original_tokenizer_class value="&lt;class 'transformers.models.whisper.tokenization_whisper_fast.WhisperTokenizerFast'>" />
 		<pad_token_id value="50257" />
+		<sentencepiece_version value="0.2.0" />
+		<skip_special_tokens value="True" />
+		<streaming_detokenizer value="False" />
+		<tiktoken_version value="0.8.0" />
+		<tokenizer_output_type value="i64" />
+		<tokenizers_version value="0.20.3" />
+		<transformers_version value="4.46.3" />
+		<use_max_padding value="False" />
+		<use_sentencepiece_backend value="False" />
+		<utf8_replace_mode />
+		<with_detokenizer value="True" />
 	</rt_info>
 </net>

openvino_tokenizer.xml

@@ -1,27 +1,27 @@
 <?xml version="1.0"?>
 <net name="tokenizer" version="11">
 	<layers>
-		<layer id="0" name="Parameter_32300" type="Parameter" version="opset1">
+		<layer id="0" name="Parameter_32640" type="Parameter" version="opset1">
 			<data shape="?" element_type="string" />
 			<output>
-				<port id="0" precision="STRING" names="Parameter_32300">
+				<port id="0" precision="STRING" names="Parameter_32640">
 					<dim>-1</dim>
 				</port>
 			</output>
 		</layer>
-		<layer id="1" name="Constant_32398" type="Const" version="opset1">
+		<layer id="1" name="Constant_32738" type="Const" version="opset1">
 			<data element_type="i32" shape="" offset="0" size="4" />
 			<output>
 				<port id="0" precision="I32" />
 			</output>
 		</layer>
-		<layer id="2" name="Constant_32399" type="Const" version="opset1">
+		<layer id="2" name="Constant_32739" type="Const" version="opset1">
 			<data element_type="i32" shape="" offset="4" size="4" />
 			<output>
 				<port id="0" precision="I32" />
 			</output>
 		</layer>
-		<layer id="3" name="Constant_32400" type="Const" version="opset1">
+		<layer id="3" name="Constant_32740" type="Const" version="opset1">
 			<data element_type="i32" shape="1" offset="8" size="4" />
 			<output>
 				<port id="0" precision="I32">
@@ -29,19 +29,19 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="4" name="Constant_32401" type="Const" version="opset1">
+		<layer id="4" name="Constant_32741" type="Const" version="opset1">
 			<data element_type="i32" shape="" offset="0" size="4" />
 			<output>
 				<port id="0" precision="I32" />
 			</output>
 		</layer>
-		<layer id="5" name="Constant_32402" type="Const" version="opset1">
+		<layer id="5" name="Constant_32742" type="Const" version="opset1">
 			<data element_type="i32" shape="" offset="4" size="4" />
 			<output>
 				<port id="0" precision="I32" />
 			</output>
 		</layer>
-		<layer id="6" name="Constant_32403" type="Const" version="opset1">
+		<layer id="6" name="Constant_32743" type="Const" version="opset1">
 			<data element_type="i32" shape="1" offset="12" size="4" />
 			<output>
 				<port id="0" precision="I32">
@@ -49,13 +49,13 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="7" name="Constant_32306" type="Const" version="opset1">
+		<layer id="7" name="Constant_32646" type="Const" version="opset1">
 			<data element_type="i64" shape="" offset="16" size="8" />
 			<output>
 				<port id="0" precision="I64" />
 			</output>
 		</layer>
-		<layer id="8" name="StringTensorUnpack_32301" type="StringTensorUnpack" version="extension">
+		<layer id="8" name="StringTensorUnpack_32641" type="StringTensorUnpack" version="extension">
 			<data mode="begins_ends" />
 			<input>
 				<port id="0" precision="STRING">
@@ -74,7 +74,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="9" name="ShapeOf_32302" type="ShapeOf" version="opset3">
+		<layer id="9" name="ShapeOf_32642" type="ShapeOf" version="opset3">
 			<data output_type="i64" />
 			<input>
 				<port id="0" precision="I32">
@@ -87,19 +87,19 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="10" name="Constant_32303" type="Const" version="opset1">
+		<layer id="10" name="Constant_32643" type="Const" version="opset1">
 			<data element_type="i64" shape="" offset="16" size="8" />
 			<output>
 				<port id="0" precision="I64" />
 			</output>
 		</layer>
-		<layer id="11" name="Constant_32304" type="Const" version="opset1">
+		<layer id="11" name="Constant_32644" type="Const" version="opset1">
 			<data element_type="i64" shape="" offset="16" size="8" />
 			<output>
 				<port id="0" precision="I64" />
 			</output>
 		</layer>
-		<layer id="12" name="Gather_32305" type="Gather" version="opset8">
+		<layer id="12" name="Gather_32645" type="Gather" version="opset8">
 			<data batch_dims="0" />
 			<input>
 				<port id="0" precision="I64">
@@ -112,13 +112,13 @@
 				<port id="3" precision="I64" />
 			</output>
 		</layer>
-		<layer id="13" name="Constant_32307" type="Const" version="opset1">
+		<layer id="13" name="Constant_32647" type="Const" version="opset1">
 			<data element_type="i64" shape="" offset="24" size="8" />
 			<output>
 				<port id="0" precision="I64" />
 			</output>
 		</layer>
-		<layer id="14" name="Range_32308" type="Range" version="opset4">
+		<layer id="14" name="Range_32648" type="Range" version="opset4">
 			<data output_type="i32" />
 			<input>
 				<port id="0" precision="I64" />
@@ -131,19 +131,19 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="15" name="Constant_32309" type="Const" version="opset1">
+		<layer id="15" name="Constant_32649" type="Const" version="opset1">
 			<data element_type="i64" shape="" offset="24" size="8" />
 			<output>
 				<port id="0" precision="I64" />
 			</output>
 		</layer>
-		<layer id="16" name="Constant_32310" type="Const" version="opset1">
+		<layer id="16" name="Constant_32650" type="Const" version="opset1">
 			<data element_type="i64" shape="" offset="24" size="8" />
 			<output>
 				<port id="0" precision="I64" />
 			</output>
 		</layer>
-		<layer id="17" name="Add_32311" type="Add" version="opset1">
+		<layer id="17" name="Add_32651" type="Add" version="opset1">
 			<data auto_broadcast="numpy" />
 			<input>
 				<port id="0" precision="I64" />
@@ -153,13 +153,13 @@
 				<port id="2" precision="I64" />
 			</output>
 		</layer>
-		<layer id="18" name="Constant_32312" type="Const" version="opset1">
+		<layer id="18" name="Constant_32652" type="Const" version="opset1">
 			<data element_type="i64" shape="" offset="24" size="8" />
 			<output>
 				<port id="0" precision="I64" />
 			</output>
 		</layer>
-		<layer id="19" name="Range_32313" type="Range" version="opset4">
+		<layer id="19" name="Range_32653" type="Range" version="opset4">
 			<data output_type="i32" />
 			<input>
 				<port id="0" precision="I64" />
@@ -172,7 +172,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="20" name="Constant_32375" type="Const" version="opset1">
+		<layer id="20" name="Constant_32715" type="Const" version="opset1">
 			<data element_type="u8" shape="26491" offset="32" size="26491" />
 			<output>
 				<port id="0" precision="U8">
@@ -180,7 +180,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="21" name="SpecialTokensSplit_32376" type="SpecialTokensSplit" version="extension">
+		<layer id="21" name="SpecialTokensSplit_32716" type="SpecialTokensSplit" version="extension">
 			<input>
 				<port id="0" precision="I32">
 					<dim>-1</dim>
@@ -222,7 +222,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="22" name="Constant_32378" type="Const" version="opset1">
+		<layer id="22" name="Constant_32718" type="Const" version="opset1">
 			<data element_type="u8" shape="64" offset="26523" size="64" />
 			<output>
 				<port id="0" precision="U8">
@@ -230,7 +230,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="23" name="RegexSplit_32379" type="RegexSplit" version="extension">
+		<layer id="23" name="RegexSplit_32719" type="RegexSplit" version="extension">
 			<data behaviour="isolate" invert="false" max_splits="-1" />
 			<input>
 				<port id="0" precision="I32">
@@ -276,7 +276,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="24" name="Constant_32381" type="Const" version="opset1">
+		<layer id="24" name="Constant_32721" type="Const" version="opset1">
 			<data element_type="u8" shape="528257" offset="26587" size="528257" />
 			<output>
 				<port id="0" precision="U8">
@@ -284,7 +284,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="25" name="StringTensorUnpack_32382" type="StringTensorUnpack" version="extension">
+		<layer id="25" name="StringTensorUnpack_32722" type="StringTensorUnpack" version="extension">
 			<data mode="begins_ends" />
 			<input>
 				<port id="0" precision="U8">
@@ -303,7 +303,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="26" name="Constant_32387" type="Const" version="opset1">
+		<layer id="26" name="Constant_32727" type="Const" version="opset1">
 			<data element_type="u8" shape="369958" offset="554844" size="369958" />
 			<output>
 				<port id="0" precision="U8">
@@ -311,7 +311,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="27" name="StringTensorUnpack_32388" type="StringTensorUnpack" version="extension">
+		<layer id="27" name="StringTensorUnpack_32728" type="StringTensorUnpack" version="extension">
 			<data mode="begins_ends" />
 			<input>
 				<port id="0" precision="U8">
@@ -330,7 +330,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="28" name="Constant_32390" type="Const" version="opset1">
+		<layer id="28" name="Constant_32730" type="Const" version="opset1">
 			<data element_type="u8" shape="336859" offset="924802" size="336859" />
 			<output>
 				<port id="0" precision="U8">
@@ -338,7 +338,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="29" name="StringTensorUnpack_32391" type="StringTensorUnpack" version="extension">
+		<layer id="29" name="StringTensorUnpack_32731" type="StringTensorUnpack" version="extension">
 			<data mode="begins_ends" />
 			<input>
 				<port id="0" precision="U8">
@@ -357,7 +357,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="30" name="Constant_32384" type="Const" version="opset1">
+		<layer id="30" name="Constant_32724" type="Const" version="opset1">
 			<data element_type="u8" shape="20172" offset="1261661" size="20172" />
 			<output>
 				<port id="0" precision="U8">
@@ -365,7 +365,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="31" name="StringTensorUnpack_32385" type="StringTensorUnpack" version="extension">
+		<layer id="31" name="StringTensorUnpack_32725" type="StringTensorUnpack" version="extension">
 			<data mode="begins_ends" />
 			<input>
 				<port id="0" precision="U8">
@@ -384,7 +384,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="32" name="Constant_32392" type="Const" version="opset1">
+		<layer id="32" name="Constant_32732" type="Const" version="opset1">
 			<data element_type="i32" shape="1609" offset="1281833" size="6436" />
 			<output>
 				<port id="0" precision="I32">
@@ -392,7 +392,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="33" name="BPETokenizer_32393" type="BPETokenizer" version="extension">
+		<layer id="33" name="BPETokenizer_32733" type="BPETokenizer" version="extension">
 			<data unk_token="" fuse_unk="false" suffix_indicator="" end_suffix="" byte_fallback="false" cache_capacity="20000" />
 			<input>
 				<port id="0" precision="I32">
@@ -462,7 +462,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="34" name="Subtract_32394" type="Subtract" version="opset1">
+		<layer id="34" name="Subtract_32734" type="Subtract" version="opset1">
 			<data auto_broadcast="numpy" />
 			<input>
 				<port id="0" precision="I32">
@@ -478,13 +478,13 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="35" name="Constant_32395" type="Const" version="opset1">
+		<layer id="35" name="Constant_32735" type="Const" version="opset1">
 			<data element_type="i32" shape="" offset="1288269" size="4" />
 			<output>
 				<port id="0" precision="I32" />
 			</output>
 		</layer>
-		<layer id="36" name="Minimum_32396" type="Minimum" version="opset1">
+		<layer id="36" name="Minimum_32736" type="Minimum" version="opset1">
 			<data auto_broadcast="numpy" />
 			<input>
 				<port id="0" precision="I32">
@@ -498,7 +498,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="37" name="Add_32397" type="Add" version="opset1">
+		<layer id="37" name="Add_32737" type="Add" version="opset1">
 			<data auto_broadcast="numpy" />
 			<input>
 				<port id="0" precision="I32">
@@ -514,19 +514,19 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="38" name="Constant_32404" type="Const" version="opset1">
+		<layer id="38" name="Constant_32744" type="Const" version="opset1">
 			<data element_type="i32" shape="" offset="0" size="4" />
 			<output>
 				<port id="0" precision="I32" />
 			</output>
 		</layer>
-		<layer id="39" name="Constant_32405" type="Const" version="opset1">
+		<layer id="39" name="Constant_32745" type="Const" version="opset1">
 			<data element_type="i32" shape="" offset="4" size="4" />
 			<output>
 				<port id="0" precision="I32" />
 			</output>
 		</layer>
-		<layer id="40" name="Constant_32406" type="Const" version="opset1">
+		<layer id="40" name="Constant_32746" type="Const" version="opset1">
 			<data element_type="i32" shape="1" offset="1288273" size="4" />
 			<output>
 				<port id="0" precision="I32">
@@ -534,7 +534,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="41" name="Constant_32407" type="Const" version="opset1">
+		<layer id="41" name="Constant_32747" type="Const" version="opset1">
 			<data element_type="i32" shape="4" offset="1288277" size="16" />
 			<output>
 				<port id="0" precision="I32">
@@ -542,7 +542,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="42" name="CombineSegments_32408" type="CombineSegments" version="extension">
+		<layer id="42" name="CombineSegments_32748" type="CombineSegments" version="extension">
 			<input>
 				<port id="0" precision="I32" />
 				<port id="1" precision="I32" />
@@ -593,7 +593,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="43" name="Subtract_32409" type="Subtract" version="opset1">
+		<layer id="43" name="Subtract_32749" type="Subtract" version="opset1">
 			<data auto_broadcast="numpy" />
 			<input>
 				<port id="0" precision="I32">
@@ -609,13 +609,13 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="44" name="Constant_32410" type="Const" version="opset1">
+		<layer id="44" name="Constant_32750" type="Const" version="opset1">
 			<data element_type="i32" shape="" offset="0" size="4" />
 			<output>
 				<port id="0" precision="I32" />
 			</output>
 		</layer>
-		<layer id="45" name="ReduceMax_32411" type="ReduceMax" version="opset1">
+		<layer id="45" name="ReduceMax_32751" type="ReduceMax" version="opset1">
 			<data keep_dims="false" />
 			<input>
 				<port id="0" precision="I32">
@@ -627,13 +627,13 @@
 				<port id="2" precision="I32" />
 			</output>
 		</layer>
-		<layer id="46" name="Constant_32412" type="Const" version="opset1">
+		<layer id="46" name="Constant_32752" type="Const" version="opset1">
 			<data element_type="i32" shape="" offset="1288273" size="4" />
 			<output>
 				<port id="0" precision="I32" />
 			</output>
 		</layer>
-		<layer id="47" name="RaggedToDense_32413" type="RaggedToDense" version="extension">
+		<layer id="47" name="RaggedToDense_32753" type="RaggedToDense" version="extension">
 			<data pad_right="true" />
 			<input>
 				<port id="0" precision="I32">
@@ -659,7 +659,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="48" name="Convert_32414" type="Convert" version="opset1">
+		<layer id="48" name="Convert_32754" type="Convert" version="opset1">
 			<data destination_type="i32" />
 			<input>
 				<port id="0" precision="BOOL">
@@ -674,7 +674,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="49" name="Convert_32414" type="Convert" version="opset1">
+		<layer id="49" name="Convert_32754" type="Convert" version="opset1">
 			<data destination_type="i64" />
 			<input>
 				<port id="0" precision="I32">
@@ -689,7 +689,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="51" name="RaggedToDense_32413.0" type="Convert" version="opset1">
+		<layer id="51" name="RaggedToDense_32753.0" type="Convert" version="opset1">
 			<data destination_type="i64" />
 			<input>
 				<port id="0" precision="I32">
@@ -704,7 +704,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="52" name="Result_32417" type="Result" version="opset1">
+		<layer id="52" name="Result_32757" type="Result" version="opset1">
 			<input>
 				<port id="0" precision="I64">
 					<dim>-1</dim>
@@ -712,7 +712,7 @@
 				</port>
 			</input>
 		</layer>
-		<layer id="50" name="Result_32419" type="Result" version="opset1">
+		<layer id="50" name="Result_32759" type="Result" version="opset1">
 			<input>
 				<port id="0" precision="I64">
 					<dim>-1</dim>
@@ -805,9 +805,29 @@
 		<edge from-layer="51" from-port="1" to-layer="52" to-port="0" />
 	</edges>
 	<rt_info>
+		<add_attention_mask value="True" />
+		<add_prefix_space />
+		<add_special_tokens value="True" />
 		<bos_token_id value="50257" />
+		<clean_up_tokenization_spaces />
+		<detokenizer_input_type value="i64" />
 		<eos_token_id value="50257" />
+		<handle_special_tokens_with_re />
+		<number_of_inputs value="1" />
+		<openvino_tokenizers_version value="2024.5.0.0" />
+		<openvino_version value="2024.5.0" />
 		<original_tokenizer_class value="&lt;class 'transformers.models.whisper.tokenization_whisper_fast.WhisperTokenizerFast'>" />
 		<pad_token_id value="50257" />
+		<sentencepiece_version value="0.2.0" />
+		<skip_special_tokens value="True" />
+		<streaming_detokenizer value="False" />
+		<tiktoken_version value="0.8.0" />
+		<tokenizer_output_type value="i64" />
+		<tokenizers_version value="0.20.3" />
+		<transformers_version value="4.46.3" />
+		<use_max_padding value="False" />
+		<use_sentencepiece_backend value="False" />
+		<utf8_replace_mode />
+		<with_detokenizer value="True" />
 	</rt_info>
 </net>