Upload folder using huggingface_hub

README.md

@@ -84,7 +84,6 @@ import openvino_genai as ov_genai
 
 device = "CPU"
 pipe = ov_genai.LLMPipeline(model_path, device)
-pipe.get_tokenizer().set_chat_template(pipe.get_tokenizer().chat_template)
 print(pipe.generate("What is OpenVINO?", max_length=200))
 ```
 

config.json

@@ -1,5 +1,4 @@
 {
-  "_attn_implementation_autoset": true,
   "architectures": [
     "Qwen3ForCausalLM"
   ],

openvino_config.json

@@ -0,0 +1,27 @@
+{
+  "dtype": "int4",
+  "input_info": null,
+  "optimum_version": "1.25.3",
+  "quantization_config": {
+    "all_layers": null,
+    "backup_precision": null,
+    "bits": 4,
+    "dataset": "wikitext2",
+    "dtype": "int4",
+    "gptq": null,
+    "group_size": 128,
+    "ignored_scope": null,
+    "lora_correction": null,
+    "num_samples": null,
+    "processor": null,
+    "quant_method": "default",
+    "ratio": 1.0,
+    "scale_estimation": true,
+    "sensitivity_metric": null,
+    "sym": false,
+    "tokenizer": null,
+    "trust_remote_code": false
+  },
+  "save_onnx_model": false,
+  "transformers_version": "4.51.3"
+}

openvino_detokenizer.xml

@@ -1,16 +1,16 @@
 <?xml version="1.0"?>
 <net name="detokenizer" version="11">
 	<layers>
-		<layer id="0" name="Parameter_110822" type="Parameter" version="opset1">
+		<layer id="0" name="Parameter_1381232" type="Parameter" version="opset1">
 			<data shape="?,?" element_type="i64" />
 			<output>
-				<port id="0" precision="I64" names="Parameter_110822">
+				<port id="0" precision="I64" names="Parameter_1381232">
 					<dim>-1</dim>
 					<dim>-1</dim>
 				</port>
 			</output>
 		</layer>
-		<layer id="1" name="Convert_110992" type="Convert" version="opset1">
+		<layer id="1" name="Convert_1381422" 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_110824" type="Const" version="opset1">
+		<layer id="2" name="Constant_1381234" type="Const" version="opset1">
 			<data element_type="i32" shape="151669" offset="0" size="606676" />
 			<output>
 				<port id="0" precision="I32">
@@ -33,7 +33,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="3" name="Constant_110826" type="Const" version="opset1">
+		<layer id="3" name="Constant_1381236" type="Const" version="opset1">
 			<data element_type="i32" shape="151669" offset="606676" size="606676" />
 			<output>
 				<port id="0" precision="I32">
@@ -41,7 +41,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="4" name="Constant_110828" type="Const" version="opset1">
+		<layer id="4" name="Constant_1381238" type="Const" version="opset1">
 			<data element_type="u8" shape="976309" offset="1213352" size="976309" />
 			<output>
 				<port id="0" precision="U8">
@@ -49,7 +49,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="5" name="Slice_110833" type="Const" version="opset1">
+		<layer id="5" name="Slice_1381243" type="Const" version="opset1">
 			<data element_type="i32" shape="14" offset="2189661" size="56" />
 			<output>
 				<port id="0" precision="I32">
@@ -57,7 +57,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="6" name="VocabDecoder_110835" type="VocabDecoder" version="extension">
+		<layer id="6" name="VocabDecoder_1381245" type="VocabDecoder" version="extension">
 			<data skip_tokens="" />
 			<input>
 				<port id="0" precision="I32">
@@ -95,7 +95,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="7" name="FuzeRagged_110836" type="FuzeRagged" version="extension">
+		<layer id="7" name="FuzeRagged_1381246" type="FuzeRagged" version="extension">
 			<input>
 				<port id="0" precision="I32">
 					<dim>-1</dim>
@@ -119,7 +119,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="8" name="UTF8Validate_110837" type="UTF8Validate" version="extension">
+		<layer id="8" name="UTF8Validate_1381247" type="UTF8Validate" version="extension">
 			<data replace_mode="true" />
 			<input>
 				<port id="0" precision="I32">
@@ -144,7 +144,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="9" name="StringTensorPack_110838" type="StringTensorPack" version="opset15">
+		<layer id="9" name="StringTensorPack_1381248" type="StringTensorPack" version="opset15">
 			<input>
 				<port id="0" precision="I32">
 					<dim>-1</dim>
@@ -157,12 +157,12 @@
 				</port>
 			</input>
 			<output>
-				<port id="3" precision="STRING" names="Result_110839,string_output">
+				<port id="3" precision="STRING" names="Result_1381249,string_output">
 					<dim>-1</dim>
 				</port>
 			</output>
 		</layer>
-		<layer id="10" name="Result_110839" type="Result" version="opset1" output_names="Result_110839,string_output">
+		<layer id="10" name="Result_1381249" type="Result" version="opset1" output_names="Result_1381249,string_output">
 			<input>
 				<port id="0" precision="STRING">
 					<dim>-1</dim>
@@ -193,15 +193,15 @@
 		<add_attention_mask value="True" />
 		<add_prefix_space />
 		<add_special_tokens value="True" />
-		<chat_template value="{%- if tools %}&#10;    {{- '&lt;|im_start|>system\n' }}&#10;    {%- if messages[0].role == 'system' %}&#10;        {{- messages[0].content + '\n\n' }}&#10;    {%- endif %}&#10;    {{- &quot;# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within &lt;tools>&lt;/tools> XML tags:\n&lt;tools>&quot; }}&#10;    {%- for tool in tools %}&#10;        {{- &quot;\n&quot; }}&#10;        {{- tool | tojson }}&#10;    {%- endfor %}&#10;    {{- &quot;\n&lt;/tools>\n\nFor each function call, return a json object with function name and arguments within &lt;tool_call>&lt;/tool_call> XML tags:\n&lt;tool_call>\n{\&quot;name\&quot;: &lt;function-name>, \&quot;arguments\&quot;: &lt;args-json-object>}\n&lt;/tool_call>&lt;|im_end|>\n&quot; }}&#10;{%- else %}&#10;    {%- if messages[0].role == 'system' %}&#10;        {{- '&lt;|im_start|>system\n' + messages[0].content + '&lt;|im_end|>\n' }}&#10;    {%- endif %}&#10;{%- endif %}&#10;{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}&#10;{%- for message in messages[::-1] %}&#10;    {%- set index = (messages|length - 1) - loop.index0 %}&#10;    {%- if ns.multi_step_tool and message.role == &quot;user&quot; and not(message.content.startswith('&lt;tool_response>') and message.content.endswith('&lt;/tool_response>')) %}&#10;        {%- set ns.multi_step_tool = false %}&#10;        {%- set ns.last_query_index = index %}&#10;    {%- endif %}&#10;{%- endfor %}&#10;{%- for message in messages %}&#10;    {%- if (message.role == &quot;user&quot;) or (message.role == &quot;system&quot; and not loop.first) %}&#10;        {{- '&lt;|im_start|>' + message.role + '\n' + message.content + '&lt;|im_end|>' + '\n' }}&#10;    {%- elif message.role == &quot;assistant&quot; %}&#10;        {%- set content = message.content %}&#10;        {%- set reasoning_content = '' %}&#10;        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}&#10;            {%- set reasoning_content = message.reasoning_content %}&#10;        {%- else %}&#10;            {%- if '&lt;/think>' in message.content %}&#10;                {%- set content = message.content.split('&lt;/think>')[-1].lstrip('\n') %}&#10;                {%- set reasoning_content = message.content.split('&lt;/think>')[0].rstrip('\n').split('&lt;think>')[-1].lstrip('\n') %}&#10;            {%- endif %}&#10;        {%- endif %}&#10;        {%- if loop.index0 > ns.last_query_index %}&#10;            {%- if loop.last or (not loop.last and reasoning_content) %}&#10;                {{- '&lt;|im_start|>' + message.role + '\n&lt;think>\n' + reasoning_content.strip('\n') + '\n&lt;/think>\n\n' + content.lstrip('\n') }}&#10;            {%- else %}&#10;                {{- '&lt;|im_start|>' + message.role + '\n' + content }}&#10;            {%- endif %}&#10;        {%- else %}&#10;            {{- '&lt;|im_start|>' + message.role + '\n' + content }}&#10;        {%- endif %}&#10;        {%- if message.tool_calls %}&#10;            {%- for tool_call in message.tool_calls %}&#10;                {%- if (loop.first and content) or (not loop.first) %}&#10;                    {{- '\n' }}&#10;                {%- endif %}&#10;                {%- if tool_call.function %}&#10;                    {%- set tool_call = tool_call.function %}&#10;                {%- endif %}&#10;                {{- '&lt;tool_call>\n{&quot;name&quot;: &quot;' }}&#10;                {{- tool_call.name }}&#10;                {{- '&quot;, &quot;arguments&quot;: ' }}&#10;                {%- if tool_call.arguments is string %}&#10;                    {{- tool_call.arguments }}&#10;                {%- else %}&#10;                    {{- tool_call.arguments | tojson }}&#10;                {%- endif %}&#10;                {{- '}\n&lt;/tool_call>' }}&#10;            {%- endfor %}&#10;        {%- endif %}&#10;        {{- '&lt;|im_end|>\n' }}&#10;    {%- elif message.role == &quot;tool&quot; %}&#10;        {%- if loop.first or (messages[loop.index0 - 1].role != &quot;tool&quot;) %}&#10;            {{- '&lt;|im_start|>user' }}&#10;        {%- endif %}&#10;        {{- '\n&lt;tool_response>\n' }}&#10;        {{- message.content }}&#10;        {{- '\n&lt;/tool_response>' }}&#10;        {%- if loop.last or (messages[loop.index0 + 1].role != &quot;tool&quot;) %}&#10;            {{- '&lt;|im_end|>\n' }}&#10;        {%- endif %}&#10;    {%- endif %}&#10;{%- endfor %}&#10;{%- if add_generation_prompt %}&#10;    {{- '&lt;|im_start|>assistant\n' }}&#10;    {%- if enable_thinking is defined and enable_thinking is false %}&#10;        {{- '&lt;think>\n\n&lt;/think>\n\n' }}&#10;    {%- endif %}&#10;{%- endif %}" />
+		<chat_template value="{%- if tools %}&#10;    {{- '&lt;|im_start|>system\n' }}&#10;    {%- if messages[0].role == 'system' %}&#10;        {{- messages[0].content + '\n\n' }}&#10;    {%- endif %}&#10;    {{- &quot;# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within &lt;tools>&lt;/tools> XML tags:\n&lt;tools>&quot; }}&#10;    {%- for tool in tools %}&#10;        {{- &quot;\n&quot; }}&#10;        {{- tool | tojson }}&#10;    {%- endfor %}&#10;    {{- &quot;\n&lt;/tools>\n\nFor each function call, return a json object with function name and arguments within &lt;tool_call>&lt;/tool_call> XML tags:\n&lt;tool_call>\n{\&quot;name\&quot;: &lt;function-name>, \&quot;arguments\&quot;: &lt;args-json-object>}\n&lt;/tool_call>&lt;|im_end|>\n&quot; }}&#10;{%- else %}&#10;    {%- if messages[0].role == 'system' %}&#10;        {{- '&lt;|im_start|>system\n' + messages[0].content + '&lt;|im_end|>\n' }}&#10;    {%- endif %}&#10;{%- endif %}&#10;{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}&#10;{%- for message in messages[::-1] %}&#10;    {%- set index = (messages|length - 1) - loop.index0 %}&#10;    {%- if ns.multi_step_tool and message.role == &quot;user&quot; and message.content is string and not(message.content.startswith('&lt;tool_response>') and message.content.endswith('&lt;/tool_response>')) %}&#10;        {%- set ns.multi_step_tool = false %}&#10;        {%- set ns.last_query_index = index %}&#10;    {%- endif %}&#10;{%- endfor %}&#10;{%- for message in messages %}&#10;    {%- if message.content is string %}&#10;        {%- set content = message.content %}&#10;    {%- else %}&#10;        {%- set content = '' %}&#10;    {%- endif %}&#10;    {%- if (message.role == &quot;user&quot;) or (message.role == &quot;system&quot; and not loop.first) %}&#10;        {{- '&lt;|im_start|>' + message.role + '\n' + content + '&lt;|im_end|>' + '\n' }}&#10;    {%- elif message.role == &quot;assistant&quot; %}&#10;        {%- set reasoning_content = '' %}&#10;        {%- if message.reasoning_content is string %}&#10;            {%- set reasoning_content = message.reasoning_content %}&#10;        {%- else %}&#10;            {%- if '&lt;/think>' in content %}&#10;                {%- set reasoning_content = content.split('&lt;/think>')[0].rstrip('\n').split('&lt;think>')[-1].lstrip('\n') %}&#10;                {%- set content = content.split('&lt;/think>')[-1].lstrip('\n') %}&#10;            {%- endif %}&#10;        {%- endif %}&#10;        {%- if loop.index0 > ns.last_query_index %}&#10;            {%- if loop.last or (not loop.last and reasoning_content) %}&#10;                {{- '&lt;|im_start|>' + message.role + '\n&lt;think>\n' + reasoning_content.strip('\n') + '\n&lt;/think>\n\n' + content.lstrip('\n') }}&#10;            {%- else %}&#10;                {{- '&lt;|im_start|>' + message.role + '\n' + content }}&#10;            {%- endif %}&#10;        {%- else %}&#10;            {{- '&lt;|im_start|>' + message.role + '\n' + content }}&#10;        {%- endif %}&#10;        {%- if message.tool_calls %}&#10;            {%- for tool_call in message.tool_calls %}&#10;                {%- if (loop.first and content) or (not loop.first) %}&#10;                    {{- '\n' }}&#10;                {%- endif %}&#10;                {%- if tool_call.function %}&#10;                    {%- set tool_call = tool_call.function %}&#10;                {%- endif %}&#10;                {{- '&lt;tool_call>\n{&quot;name&quot;: &quot;' }}&#10;                {{- tool_call.name }}&#10;                {{- '&quot;, &quot;arguments&quot;: ' }}&#10;                {%- if tool_call.arguments is string %}&#10;                    {{- tool_call.arguments }}&#10;                {%- else %}&#10;                    {{- tool_call.arguments | tojson }}&#10;                {%- endif %}&#10;                {{- '}\n&lt;/tool_call>' }}&#10;            {%- endfor %}&#10;        {%- endif %}&#10;        {{- '&lt;|im_end|>\n' }}&#10;    {%- elif message.role == &quot;tool&quot; %}&#10;        {%- if loop.first or (messages[loop.index0 - 1].role != &quot;tool&quot;) %}&#10;            {{- '&lt;|im_start|>user' }}&#10;        {%- endif %}&#10;        {{- '\n&lt;tool_response>\n' }}&#10;        {{- content }}&#10;        {{- '\n&lt;/tool_response>' }}&#10;        {%- if loop.last or (messages[loop.index0 + 1].role != &quot;tool&quot;) %}&#10;            {{- '&lt;|im_end|>\n' }}&#10;        {%- endif %}&#10;    {%- endif %}&#10;{%- endfor %}&#10;{%- if add_generation_prompt %}&#10;    {{- '&lt;|im_start|>assistant\n' }}&#10;    {%- if enable_thinking is defined and enable_thinking is false %}&#10;        {{- '&lt;think>\n\n&lt;/think>\n\n' }}&#10;    {%- endif %}&#10;{%- endif %}" />
 		<clean_up_tokenization_spaces />
 		<detokenizer_input_type value="i64" />
 		<eos_token_id value="151645" />
 		<handle_special_tokens_with_re />
 		<max_length />
 		<number_of_inputs value="1" />
-		<openvino_tokenizers_version value="2025.1.0.0-523-710ddf14de8" />
-		<openvino_version value="2025.1.0-18503-6fec06580ab-releases/2025/1" />
+		<openvino_tokenizers_version value="2025.2.0.0-566-85be884a69f" />
+		<openvino_version value="2025.2.0-19140-c01cd93e24d-releases/2025/2" />
 		<original_post_processor_template value="{&quot;type&quot;: &quot;ByteLevel&quot;, &quot;add_prefix_space&quot;: false, &quot;trim_offsets&quot;: false, &quot;use_regex&quot;: false}" />
 		<original_tokenizer_class value="&lt;class 'transformers.models.qwen2.tokenization_qwen2_fast.Qwen2TokenizerFast'>" />
 		<pad_token_id value="151643" />

openvino_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b3204fd8bb2963f386d3978948d29fbce26a6279c49046d63f836658c7c289d6
-size 5531479900
+oid sha256:c4824333ff1917ecaae5cce0eb6f0e572608106cbbb9acb9c8538d5a57e7dd20
+size 4855348060

openvino_tokenizer.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1e31f7285e95a4d08f1e5516c5099f96e169253100b4919db936990fa1923395
-size 5588868
+oid sha256:7f5ad873902d49091f9698d375d992b80f988fc077014a24ba5c78336bf03573
+size 5588885

openvino_tokenizer.xml

@@ -1,21 +1,21 @@
 <?xml version="1.0"?>
 <net name="tokenizer" version="11">
 	<layers>
-		<layer id="0" name="Parameter_110703" type="Parameter" version="opset1">
+		<layer id="0" name="Parameter_1381111" type="Parameter" version="opset1">
 			<data shape="?" element_type="string" />
 			<output>
-				<port id="0" precision="STRING" names="Parameter_110703">
+				<port id="0" precision="STRING" names="Parameter_1381111">
 					<dim>-1</dim>
 				</port>
 			</output>
 		</layer>
-		<layer id="1" name="Constant_110709" type="Const" version="opset1">
+		<layer id="1" name="Constant_1381117" type="Const" version="opset1">
 			<data element_type="i64" shape="" offset="0" size="8" />
 			<output>
 				<port id="0" precision="I64" />
 			</output>
 		</layer>
-		<layer id="2" name="StringTensorUnpack_110704" type="StringTensorUnpack" version="opset15">
+		<layer id="2" name="StringTensorUnpack_1381112" type="StringTensorUnpack" version="opset15">
 			<input>
 				<port id="0" precision="STRING">
 					<dim>-1</dim>
@@ -33,7 +33,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="3" name="ShapeOf_110705" type="ShapeOf" version="opset3">
+		<layer id="3" name="ShapeOf_1381113" type="ShapeOf" version="opset3">
 			<data output_type="i64" />
 			<input>
 				<port id="0" precision="I32">
@@ -46,19 +46,19 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="4" name="Constant_110706" type="Const" version="opset1">
+		<layer id="4" name="Constant_1381114" type="Const" version="opset1">
 			<data element_type="i64" shape="" offset="0" size="8" />
 			<output>
 				<port id="0" precision="I64" />
 			</output>
 		</layer>
-		<layer id="5" name="Constant_110707" type="Const" version="opset1">
+		<layer id="5" name="Constant_1381115" type="Const" version="opset1">
 			<data element_type="i64" shape="" offset="0" size="8" />
 			<output>
 				<port id="0" precision="I64" />
 			</output>
 		</layer>
-		<layer id="6" name="Gather_110708" type="Gather" version="opset8">
+		<layer id="6" name="Gather_1381116" type="Gather" version="opset8">
 			<data batch_dims="0" />
 			<input>
 				<port id="0" precision="I64">
@@ -71,13 +71,13 @@
 				<port id="3" precision="I64" />
 			</output>
 		</layer>
-		<layer id="7" name="Constant_110710" type="Const" version="opset1">
+		<layer id="7" name="Constant_1381118" type="Const" version="opset1">
 			<data element_type="i64" shape="" offset="8" size="8" />
 			<output>
 				<port id="0" precision="I64" />
 			</output>
 		</layer>
-		<layer id="8" name="Range_110711" type="Range" version="opset4">
+		<layer id="8" name="Range_1381119" type="Range" version="opset4">
 			<data output_type="i32" />
 			<input>
 				<port id="0" precision="I64" />
@@ -90,19 +90,19 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="9" name="Constant_110712" type="Const" version="opset1">
+		<layer id="9" name="Constant_1381120" type="Const" version="opset1">
 			<data element_type="i64" shape="" offset="8" size="8" />
 			<output>
 				<port id="0" precision="I64" />
 			</output>
 		</layer>
-		<layer id="10" name="Constant_110713" type="Const" version="opset1">
+		<layer id="10" name="Constant_1381121" type="Const" version="opset1">
 			<data element_type="i64" shape="" offset="8" size="8" />
 			<output>
 				<port id="0" precision="I64" />
 			</output>
 		</layer>
-		<layer id="11" name="Add_110714" type="Add" version="opset1">
+		<layer id="11" name="Add_1381122" type="Add" version="opset1">
 			<data auto_broadcast="numpy" />
 			<input>
 				<port id="0" precision="I64" />
@@ -112,13 +112,13 @@
 				<port id="2" precision="I64" />
 			</output>
 		</layer>
-		<layer id="12" name="Constant_110715" type="Const" version="opset1">
+		<layer id="12" name="Constant_1381123" type="Const" version="opset1">
 			<data element_type="i64" shape="" offset="8" size="8" />
 			<output>
 				<port id="0" precision="I64" />
 			</output>
 		</layer>
-		<layer id="13" name="Range_110716" type="Range" version="opset4">
+		<layer id="13" name="Range_1381124" type="Range" version="opset4">
 			<data output_type="i32" />
 			<input>
 				<port id="0" precision="I64" />
@@ -131,7 +131,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="14" name="Constant_110778" type="Const" version="opset1">
+		<layer id="14" name="Constant_1381186" type="Const" version="opset1">
 			<data element_type="u8" shape="512" offset="16" size="512" />
 			<output>
 				<port id="0" precision="U8">
@@ -139,7 +139,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="15" name="SpecialTokensSplit_110779" type="SpecialTokensSplit" version="extension">
+		<layer id="15" name="SpecialTokensSplit_1381187" type="SpecialTokensSplit" version="extension">
 			<input>
 				<port id="0" precision="I32">
 					<dim>-1</dim>
@@ -181,7 +181,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="16" name="CharsMapNormalization_110780" type="CharsMapNormalization" version="extension">
+		<layer id="16" name="CharsMapNormalization_1381188" type="CharsMapNormalization" version="extension">
 			<data add_dummy_prefix="false" remove_extra_whitespaces="false" escape_whitespaces="false" normalization_form="nfc" case_fold="false" nmt="false" />
 			<input>
 				<port id="0" precision="I32">
@@ -212,7 +212,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="17" name="Constant_110782" type="Const" version="opset1">
+		<layer id="17" name="Constant_1381190" type="Const" version="opset1">
 			<data element_type="u8" shape="110" offset="528" size="110" />
 			<output>
 				<port id="0" precision="U8">
@@ -220,7 +220,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="18" name="RegexSplit_110783" type="RegexSplit" version="extension">
+		<layer id="18" name="RegexSplit_1381191" type="RegexSplit" version="extension">
 			<data behaviour="isolate" invert="false" max_splits="-1" />
 			<input>
 				<port id="0" precision="I32">
@@ -266,7 +266,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="19" name="Constant_110785" type="Const" version="opset1">
+		<layer id="19" name="Constant_1381193" type="Const" version="opset1">
 			<data element_type="i32" shape="151669" offset="638" size="606676" />
 			<output>
 				<port id="0" precision="I32">
@@ -274,7 +274,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="20" name="Constant_110787" type="Const" version="opset1">
+		<layer id="20" name="Constant_1381195" type="Const" version="opset1">
 			<data element_type="i32" shape="151669" offset="607314" size="606676" />
 			<output>
 				<port id="0" precision="I32">
@@ -282,7 +282,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="21" name="Constant_110789" type="Const" version="opset1">
+		<layer id="21" name="Constant_1381197" type="Const" version="opset1">
 			<data element_type="u8" shape="976309" offset="1213990" size="976309" />
 			<output>
 				<port id="0" precision="U8">
@@ -290,7 +290,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="22" name="Constant_110797" type="Const" version="opset1">
+		<layer id="22" name="Constant_1381205" type="Const" version="opset1">
 			<data element_type="i32" shape="151387" offset="2190299" size="605548" />
 			<output>
 				<port id="0" precision="I32">
@@ -298,7 +298,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="23" name="Constant_110799" type="Const" version="opset1">
+		<layer id="23" name="Constant_1381207" type="Const" version="opset1">
 			<data element_type="i32" shape="151387" offset="2795847" size="605548" />
 			<output>
 				<port id="0" precision="I32">
@@ -306,7 +306,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="24" name="Constant_110801" type="Const" version="opset1">
+		<layer id="24" name="Constant_1381209" type="Const" version="opset1">
 			<data element_type="u8" shape="491359" offset="3401395" size="491359" />
 			<output>
 				<port id="0" precision="U8">
@@ -314,7 +314,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="25" name="Constant_110803" type="Const" version="opset1">
+		<layer id="25" name="Constant_1381211" type="Const" version="opset1">
 			<data element_type="i32" shape="151387" offset="3892754" size="605548" />
 			<output>
 				<port id="0" precision="I32">
@@ -322,7 +322,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="26" name="Constant_110805" type="Const" version="opset1">
+		<layer id="26" name="Constant_1381213" type="Const" version="opset1">
 			<data element_type="i32" shape="151387" offset="4498302" size="605548" />
 			<output>
 				<port id="0" precision="I32">
@@ -330,7 +330,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="27" name="Constant_110807" type="Const" version="opset1">
+		<layer id="27" name="Constant_1381215" type="Const" version="opset1">
 			<data element_type="u8" shape="484354" offset="5103850" size="484354" />
 			<output>
 				<port id="0" precision="U8">
@@ -338,7 +338,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="28" name="Constant_110791" type="Const" version="opset1">
+		<layer id="28" name="Constant_1381199" type="Const" version="opset1">
 			<data element_type="i32" shape="26" offset="5588204" size="104" />
 			<output>
 				<port id="0" precision="I32">
@@ -346,7 +346,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="29" name="Constant_110793" type="Const" version="opset1">
+		<layer id="29" name="Constant_1381201" type="Const" version="opset1">
 			<data element_type="i32" shape="26" offset="5588308" size="104" />
 			<output>
 				<port id="0" precision="I32">
@@ -354,7 +354,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="30" name="Constant_110795" type="Const" version="opset1">
+		<layer id="30" name="Constant_1381203" type="Const" version="opset1">
 			<data element_type="u8" shape="340" offset="5588412" size="340" />
 			<output>
 				<port id="0" precision="U8">
@@ -362,7 +362,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="31" name="Constant_110808" type="Const" version="opset1">
+		<layer id="31" name="Constant_1381216" type="Const" version="opset1">
 			<data element_type="i32" shape="26" offset="5588752" size="104" />
 			<output>
 				<port id="0" precision="I32">
@@ -370,7 +370,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="32" name="BPETokenizer_110809" type="BPETokenizer" version="extension">
+		<layer id="32" name="BPETokenizer_1381217" type="BPETokenizer" version="extension">
 			<data unk_token="" fuse_unk="false" suffix_indicator="" end_suffix="" byte_fallback="false" cache_capacity="30328" />
 			<input>
 				<port id="0" precision="I32">
@@ -440,44 +440,30 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="33" name="Subtract_110810" type="Subtract" version="opset1">
-			<data auto_broadcast="numpy" />
-			<input>
-				<port id="0" precision="I32">
-					<dim>-1</dim>
-				</port>
-				<port id="1" precision="I32">
-					<dim>-1</dim>
-				</port>
-			</input>
-			<output>
-				<port id="2" precision="I32">
-					<dim>-1</dim>
-				</port>
-			</output>
-		</layer>
-		<layer id="34" name="Constant_110811" type="Const" version="opset1">
+		<layer id="33" name="Constant_1381218" type="Const" version="opset1">
 			<data element_type="i32" shape="" offset="5588856" size="4" />
 			<output>
 				<port id="0" precision="I32" />
 			</output>
 		</layer>
-		<layer id="35" name="Minimum_110812" type="Minimum" version="opset1">
-			<data auto_broadcast="numpy" />
-			<input>
-				<port id="0" precision="I32">
-					<dim>-1</dim>
+		<layer id="34" name="Constant_1381220" type="Const" version="opset1">
+			<data element_type="u8" shape="4" offset="5588860" size="4" />
+			<output>
+				<port id="0" precision="U8">
+					<dim>4</dim>
 				</port>
-				<port id="1" precision="I32" />
-			</input>
+			</output>
+		</layer>
+		<layer id="35" name="Constant_1381222" type="Const" version="opset1">
+			<data element_type="u8" shape="13" offset="5588864" size="13" />
 			<output>
-				<port id="2" precision="I32">
-					<dim>-1</dim>
+				<port id="0" precision="U8">
+					<dim>13</dim>
 				</port>
 			</output>
 		</layer>
-		<layer id="36" name="Subtract_110813" type="Subtract" version="opset1">
-			<data auto_broadcast="numpy" />
+		<layer id="36" name="Truncate_1381223" type="Truncate" version="extension">
+			<data m_num_inputs="1" />
 			<input>
 				<port id="0" precision="I32">
 					<dim>-1</dim>
@@ -485,14 +471,30 @@
 				<port id="1" precision="I32">
 					<dim>-1</dim>
 				</port>
+				<port id="2" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="3" precision="I32" />
+				<port id="4" precision="U8">
+					<dim>4</dim>
+				</port>
+				<port id="5" precision="U8">
+					<dim>13</dim>
+				</port>
 			</input>
 			<output>
-				<port id="2" precision="I32">
+				<port id="6" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="7" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="8" precision="I32">
 					<dim>-1</dim>
 				</port>
 			</output>
 		</layer>
-		<layer id="37" name="Subtract_110814" type="Subtract" version="opset1">
+		<layer id="37" name="Subtract_1381224" type="Subtract" version="opset1">
 			<data auto_broadcast="numpy" />
 			<input>
 				<port id="0" precision="I32">
@@ -508,13 +510,13 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="38" name="Constant_110815" type="Const" version="opset1">
-			<data element_type="i32" shape="" offset="5588860" size="4" />
+		<layer id="38" name="Constant_1381225" type="Const" version="opset1">
+			<data element_type="i32" shape="" offset="5588877" size="4" />
 			<output>
 				<port id="0" precision="I32" />
 			</output>
 		</layer>
-		<layer id="39" name="ReduceMax_110816" type="ReduceMax" version="opset1">
+		<layer id="39" name="ReduceMax_1381226" type="ReduceMax" version="opset1">
 			<data keep_dims="false" />
 			<input>
 				<port id="0" precision="I32">
@@ -526,13 +528,13 @@
 				<port id="2" precision="I32" />
 			</output>
 		</layer>
-		<layer id="40" name="Constant_110817" type="Const" version="opset1">
-			<data element_type="i32" shape="" offset="5588864" size="4" />
+		<layer id="40" name="Constant_1381227" type="Const" version="opset1">
+			<data element_type="i32" shape="" offset="5588881" size="4" />
 			<output>
 				<port id="0" precision="I32" />
 			</output>
 		</layer>
-		<layer id="41" name="RaggedToDense_110818" type="RaggedToDense" version="extension">
+		<layer id="41" name="RaggedToDense_1381228" type="RaggedToDense" version="extension">
 			<data pad_right="false" m_pad_max_length="false" />
 			<input>
 				<port id="0" precision="I32">
@@ -558,7 +560,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="42" name="Convert_110819" type="Convert" version="opset1">
+		<layer id="42" name="Convert_1381229" type="Convert" version="opset1">
 			<data destination_type="i32" />
 			<input>
 				<port id="0" precision="BOOL">
@@ -573,7 +575,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="43" name="Convert_110819.0" type="Convert" version="opset1">
+		<layer id="43" name="Convert_1381229.0" type="Convert" version="opset1">
 			<data destination_type="i64" />
 			<input>
 				<port id="0" precision="I32">
@@ -588,7 +590,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="45" name="RaggedToDense_110818.0" type="Convert" version="opset1">
+		<layer id="45" name="RaggedToDense_1381228.0" type="Convert" version="opset1">
 			<data destination_type="i64" />
 			<input>
 				<port id="0" precision="I32">
@@ -603,7 +605,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="46" name="Result_110820" type="Result" version="opset1" output_names="input_ids">
+		<layer id="46" name="Result_1381230" type="Result" version="opset1" output_names="input_ids">
 			<input>
 				<port id="0" precision="I64">
 					<dim>-1</dim>
@@ -611,7 +613,7 @@
 				</port>
 			</input>
 		</layer>
-		<layer id="44" name="Result_110821" type="Result" version="opset1" output_names="attention_mask">
+		<layer id="44" name="Result_1381231" type="Result" version="opset1" output_names="attention_mask">
 			<input>
 				<port id="0" precision="I64">
 					<dim>-1</dim>
@@ -654,8 +656,8 @@
 		<edge from-layer="18" from-port="11" to-layer="32" to-port="4" />
 		<edge from-layer="18" from-port="10" to-layer="32" to-port="3" />
 		<edge from-layer="18" from-port="8" to-layer="32" to-port="1" />
-		<edge from-layer="18" from-port="7" to-layer="32" to-port="0" />
 		<edge from-layer="18" from-port="9" to-layer="32" to-port="2" />
+		<edge from-layer="18" from-port="7" to-layer="32" to-port="0" />
 		<edge from-layer="19" from-port="0" to-layer="32" to-port="5" />
 		<edge from-layer="20" from-port="0" to-layer="32" to-port="6" />
 		<edge from-layer="21" from-port="0" to-layer="32" to-port="7" />
@@ -669,17 +671,17 @@
 		<edge from-layer="29" from-port="0" to-layer="32" to-port="15" />
 		<edge from-layer="30" from-port="0" to-layer="32" to-port="16" />
 		<edge from-layer="31" from-port="0" to-layer="32" to-port="17" />
-		<edge from-layer="32" from-port="19" to-layer="33" to-port="0" />
-		<edge from-layer="32" from-port="18" to-layer="33" to-port="1" />
-		<edge from-layer="32" from-port="20" to-layer="41" to-port="2" />
-		<edge from-layer="32" from-port="19" to-layer="36" to-port="0" />
-		<edge from-layer="32" from-port="19" to-layer="41" to-port="1" />
-		<edge from-layer="32" from-port="19" to-layer="37" to-port="0" />
-		<edge from-layer="33" from-port="2" to-layer="35" to-port="0" />
-		<edge from-layer="34" from-port="0" to-layer="35" to-port="1" />
-		<edge from-layer="35" from-port="2" to-layer="36" to-port="1" />
-		<edge from-layer="36" from-port="2" to-layer="37" to-port="1" />
-		<edge from-layer="36" from-port="2" to-layer="41" to-port="0" />
+		<edge from-layer="32" from-port="20" to-layer="36" to-port="2" />
+		<edge from-layer="32" from-port="18" to-layer="36" to-port="0" />
+		<edge from-layer="32" from-port="19" to-layer="36" to-port="1" />
+		<edge from-layer="33" from-port="0" to-layer="36" to-port="3" />
+		<edge from-layer="34" from-port="0" to-layer="36" to-port="4" />
+		<edge from-layer="35" from-port="0" to-layer="36" to-port="5" />
+		<edge from-layer="36" from-port="6" to-layer="41" to-port="0" />
+		<edge from-layer="36" from-port="8" to-layer="41" to-port="2" />
+		<edge from-layer="36" from-port="7" to-layer="41" to-port="1" />
+		<edge from-layer="36" from-port="6" to-layer="37" to-port="1" />
+		<edge from-layer="36" from-port="7" to-layer="37" to-port="0" />
 		<edge from-layer="37" from-port="2" to-layer="39" to-port="0" />
 		<edge from-layer="38" from-port="0" to-layer="39" to-port="1" />
 		<edge from-layer="39" from-port="2" to-layer="41" to-port="3" />
@@ -694,20 +696,19 @@
 		<add_attention_mask value="True" />
 		<add_prefix_space />
 		<add_special_tokens value="True" />
-		<chat_template value="{%- if tools %}&#10;    {{- '&lt;|im_start|>system\n' }}&#10;    {%- if messages[0].role == 'system' %}&#10;        {{- messages[0].content + '\n\n' }}&#10;    {%- endif %}&#10;    {{- &quot;# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within &lt;tools>&lt;/tools> XML tags:\n&lt;tools>&quot; }}&#10;    {%- for tool in tools %}&#10;        {{- &quot;\n&quot; }}&#10;        {{- tool | tojson }}&#10;    {%- endfor %}&#10;    {{- &quot;\n&lt;/tools>\n\nFor each function call, return a json object with function name and arguments within &lt;tool_call>&lt;/tool_call> XML tags:\n&lt;tool_call>\n{\&quot;name\&quot;: &lt;function-name>, \&quot;arguments\&quot;: &lt;args-json-object>}\n&lt;/tool_call>&lt;|im_end|>\n&quot; }}&#10;{%- else %}&#10;    {%- if messages[0].role == 'system' %}&#10;        {{- '&lt;|im_start|>system\n' + messages[0].content + '&lt;|im_end|>\n' }}&#10;    {%- endif %}&#10;{%- endif %}&#10;{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}&#10;{%- for message in messages[::-1] %}&#10;    {%- set index = (messages|length - 1) - loop.index0 %}&#10;    {%- if ns.multi_step_tool and message.role == &quot;user&quot; and not(message.content.startswith('&lt;tool_response>') and message.content.endswith('&lt;/tool_response>')) %}&#10;        {%- set ns.multi_step_tool = false %}&#10;        {%- set ns.last_query_index = index %}&#10;    {%- endif %}&#10;{%- endfor %}&#10;{%- for message in messages %}&#10;    {%- if (message.role == &quot;user&quot;) or (message.role == &quot;system&quot; and not loop.first) %}&#10;        {{- '&lt;|im_start|>' + message.role + '\n' + message.content + '&lt;|im_end|>' + '\n' }}&#10;    {%- elif message.role == &quot;assistant&quot; %}&#10;        {%- set content = message.content %}&#10;        {%- set reasoning_content = '' %}&#10;        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}&#10;            {%- set reasoning_content = message.reasoning_content %}&#10;        {%- else %}&#10;            {%- if '&lt;/think>' in message.content %}&#10;                {%- set content = message.content.split('&lt;/think>')[-1].lstrip('\n') %}&#10;                {%- set reasoning_content = message.content.split('&lt;/think>')[0].rstrip('\n').split('&lt;think>')[-1].lstrip('\n') %}&#10;            {%- endif %}&#10;        {%- endif %}&#10;        {%- if loop.index0 > ns.last_query_index %}&#10;            {%- if loop.last or (not loop.last and reasoning_content) %}&#10;                {{- '&lt;|im_start|>' + message.role + '\n&lt;think>\n' + reasoning_content.strip('\n') + '\n&lt;/think>\n\n' + content.lstrip('\n') }}&#10;            {%- else %}&#10;                {{- '&lt;|im_start|>' + message.role + '\n' + content }}&#10;            {%- endif %}&#10;        {%- else %}&#10;            {{- '&lt;|im_start|>' + message.role + '\n' + content }}&#10;        {%- endif %}&#10;        {%- if message.tool_calls %}&#10;            {%- for tool_call in message.tool_calls %}&#10;                {%- if (loop.first and content) or (not loop.first) %}&#10;                    {{- '\n' }}&#10;                {%- endif %}&#10;                {%- if tool_call.function %}&#10;                    {%- set tool_call = tool_call.function %}&#10;                {%- endif %}&#10;                {{- '&lt;tool_call>\n{&quot;name&quot;: &quot;' }}&#10;                {{- tool_call.name }}&#10;                {{- '&quot;, &quot;arguments&quot;: ' }}&#10;                {%- if tool_call.arguments is string %}&#10;                    {{- tool_call.arguments }}&#10;                {%- else %}&#10;                    {{- tool_call.arguments | tojson }}&#10;                {%- endif %}&#10;                {{- '}\n&lt;/tool_call>' }}&#10;            {%- endfor %}&#10;        {%- endif %}&#10;        {{- '&lt;|im_end|>\n' }}&#10;    {%- elif message.role == &quot;tool&quot; %}&#10;        {%- if loop.first or (messages[loop.index0 - 1].role != &quot;tool&quot;) %}&#10;            {{- '&lt;|im_start|>user' }}&#10;        {%- endif %}&#10;        {{- '\n&lt;tool_response>\n' }}&#10;        {{- message.content }}&#10;        {{- '\n&lt;/tool_response>' }}&#10;        {%- if loop.last or (messages[loop.index0 + 1].role != &quot;tool&quot;) %}&#10;            {{- '&lt;|im_end|>\n' }}&#10;        {%- endif %}&#10;    {%- endif %}&#10;{%- endfor %}&#10;{%- if add_generation_prompt %}&#10;    {{- '&lt;|im_start|>assistant\n' }}&#10;    {%- if enable_thinking is defined and enable_thinking is false %}&#10;        {{- '&lt;think>\n\n&lt;/think>\n\n' }}&#10;    {%- endif %}&#10;{%- endif %}" />
+		<chat_template value="{%- if tools %}&#10;    {{- '&lt;|im_start|>system\n' }}&#10;    {%- if messages[0].role == 'system' %}&#10;        {{- messages[0].content + '\n\n' }}&#10;    {%- endif %}&#10;    {{- &quot;# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within &lt;tools>&lt;/tools> XML tags:\n&lt;tools>&quot; }}&#10;    {%- for tool in tools %}&#10;        {{- &quot;\n&quot; }}&#10;        {{- tool | tojson }}&#10;    {%- endfor %}&#10;    {{- &quot;\n&lt;/tools>\n\nFor each function call, return a json object with function name and arguments within &lt;tool_call>&lt;/tool_call> XML tags:\n&lt;tool_call>\n{\&quot;name\&quot;: &lt;function-name>, \&quot;arguments\&quot;: &lt;args-json-object>}\n&lt;/tool_call>&lt;|im_end|>\n&quot; }}&#10;{%- else %}&#10;    {%- if messages[0].role == 'system' %}&#10;        {{- '&lt;|im_start|>system\n' + messages[0].content + '&lt;|im_end|>\n' }}&#10;    {%- endif %}&#10;{%- endif %}&#10;{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}&#10;{%- for message in messages[::-1] %}&#10;    {%- set index = (messages|length - 1) - loop.index0 %}&#10;    {%- if ns.multi_step_tool and message.role == &quot;user&quot; and message.content is string and not(message.content.startswith('&lt;tool_response>') and message.content.endswith('&lt;/tool_response>')) %}&#10;        {%- set ns.multi_step_tool = false %}&#10;        {%- set ns.last_query_index = index %}&#10;    {%- endif %}&#10;{%- endfor %}&#10;{%- for message in messages %}&#10;    {%- if message.content is string %}&#10;        {%- set content = message.content %}&#10;    {%- else %}&#10;        {%- set content = '' %}&#10;    {%- endif %}&#10;    {%- if (message.role == &quot;user&quot;) or (message.role == &quot;system&quot; and not loop.first) %}&#10;        {{- '&lt;|im_start|>' + message.role + '\n' + content + '&lt;|im_end|>' + '\n' }}&#10;    {%- elif message.role == &quot;assistant&quot; %}&#10;        {%- set reasoning_content = '' %}&#10;        {%- if message.reasoning_content is string %}&#10;            {%- set reasoning_content = message.reasoning_content %}&#10;        {%- else %}&#10;            {%- if '&lt;/think>' in content %}&#10;                {%- set reasoning_content = content.split('&lt;/think>')[0].rstrip('\n').split('&lt;think>')[-1].lstrip('\n') %}&#10;                {%- set content = content.split('&lt;/think>')[-1].lstrip('\n') %}&#10;            {%- endif %}&#10;        {%- endif %}&#10;        {%- if loop.index0 > ns.last_query_index %}&#10;            {%- if loop.last or (not loop.last and reasoning_content) %}&#10;                {{- '&lt;|im_start|>' + message.role + '\n&lt;think>\n' + reasoning_content.strip('\n') + '\n&lt;/think>\n\n' + content.lstrip('\n') }}&#10;            {%- else %}&#10;                {{- '&lt;|im_start|>' + message.role + '\n' + content }}&#10;            {%- endif %}&#10;        {%- else %}&#10;            {{- '&lt;|im_start|>' + message.role + '\n' + content }}&#10;        {%- endif %}&#10;        {%- if message.tool_calls %}&#10;            {%- for tool_call in message.tool_calls %}&#10;                {%- if (loop.first and content) or (not loop.first) %}&#10;                    {{- '\n' }}&#10;                {%- endif %}&#10;                {%- if tool_call.function %}&#10;                    {%- set tool_call = tool_call.function %}&#10;                {%- endif %}&#10;                {{- '&lt;tool_call>\n{&quot;name&quot;: &quot;' }}&#10;                {{- tool_call.name }}&#10;                {{- '&quot;, &quot;arguments&quot;: ' }}&#10;                {%- if tool_call.arguments is string %}&#10;                    {{- tool_call.arguments }}&#10;                {%- else %}&#10;                    {{- tool_call.arguments | tojson }}&#10;                {%- endif %}&#10;                {{- '}\n&lt;/tool_call>' }}&#10;            {%- endfor %}&#10;        {%- endif %}&#10;        {{- '&lt;|im_end|>\n' }}&#10;    {%- elif message.role == &quot;tool&quot; %}&#10;        {%- if loop.first or (messages[loop.index0 - 1].role != &quot;tool&quot;) %}&#10;            {{- '&lt;|im_start|>user' }}&#10;        {%- endif %}&#10;        {{- '\n&lt;tool_response>\n' }}&#10;        {{- content }}&#10;        {{- '\n&lt;/tool_response>' }}&#10;        {%- if loop.last or (messages[loop.index0 + 1].role != &quot;tool&quot;) %}&#10;            {{- '&lt;|im_end|>\n' }}&#10;        {%- endif %}&#10;    {%- endif %}&#10;{%- endfor %}&#10;{%- if add_generation_prompt %}&#10;    {{- '&lt;|im_start|>assistant\n' }}&#10;    {%- if enable_thinking is defined and enable_thinking is false %}&#10;        {{- '&lt;think>\n\n&lt;/think>\n\n' }}&#10;    {%- endif %}&#10;{%- endif %}" />
 		<clean_up_tokenization_spaces />
 		<detokenizer_input_type value="i64" />
 		<eos_token_id value="151645" />
 		<handle_special_tokens_with_re />
 		<max_length />
 		<number_of_inputs value="1" />
-		<openvino_tokenizers_version value="2025.1.0.0-523-710ddf14de8" />
-		<openvino_version value="2025.1.0-18503-6fec06580ab-releases/2025/1" />
+		<openvino_tokenizers_version value="2025.2.0.0-566-85be884a69f" />
+		<openvino_version value="2025.2.0-19140-c01cd93e24d-releases/2025/2" />
 		<original_post_processor_template value="{&quot;type&quot;: &quot;ByteLevel&quot;, &quot;add_prefix_space&quot;: false, &quot;trim_offsets&quot;: false, &quot;use_regex&quot;: false}" />
 		<original_tokenizer_class value="&lt;class 'transformers.models.qwen2.tokenization_qwen2_fast.Qwen2TokenizerFast'>" />
 		<pad_token_id value="151643" />
 		<sentencepiece_version value="0.2.0" />
-		<simplified_chat_template value="{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '&lt;|im_start|>system&#10;You are a helpful assistant.&lt;|im_end|>&#10;' }}{% endif %}{{'&lt;|im_start|>' + message['role'] + '&#10;' + message['content'] + '&lt;|im_end|>' + '&#10;'}}{% endfor %}{% if add_generation_prompt %}{{ '&lt;|im_start|>assistant&#10;' }}{% endif %}" />
 		<skip_special_tokens value="True" />
 		<streaming_detokenizer value="False" />
 		<tiktoken_version value="0.9.0" />

tokenizer_config.json

@@ -227,7 +227,7 @@
     "<|video_pad|>"
   ],
   "bos_token": null,
-  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0].role == 'system' %}\n        {{- messages[0].content + '\\n\\n' }}\n    {%- endif %}\n    {{- \"# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n    {%- if messages[0].role == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0].content + '<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}\n{%- for message in messages[::-1] %}\n    {%- set index = (messages|length - 1) - loop.index0 %}\n    {%- if ns.multi_step_tool and message.role == \"user\" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}\n        {%- set ns.multi_step_tool = false %}\n        {%- set ns.last_query_index = index %}\n    {%- endif %}\n{%- endfor %}\n{%- for message in messages %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) %}\n        {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {%- set content = message.content %}\n        {%- set reasoning_content = '' %}\n        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}\n            {%- set reasoning_content = message.reasoning_content %}\n        {%- else %}\n            {%- if '</think>' in message.content %}\n                {%- set content = message.content.split('</think>')[-1].lstrip('\\n') %}\n                {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\\n').split('<think>')[-1].lstrip('\\n') %}\n            {%- endif %}\n        {%- endif %}\n        {%- if loop.index0 > ns.last_query_index %}\n            {%- if loop.last or (not loop.last and reasoning_content) %}\n                {{- '<|im_start|>' + message.role + '\\n<think>\\n' + reasoning_content.strip('\\n') + '\\n</think>\\n\\n' + content.lstrip('\\n') }}\n            {%- else %}\n                {{- '<|im_start|>' + message.role + '\\n' + content }}\n            {%- endif %}\n        {%- else %}\n            {{- '<|im_start|>' + message.role + '\\n' + content }}\n        {%- endif %}\n        {%- if message.tool_calls %}\n            {%- for tool_call in message.tool_calls %}\n                {%- if (loop.first and content) or (not loop.first) %}\n                    {{- '\\n' }}\n                {%- endif %}\n                {%- if tool_call.function %}\n                    {%- set tool_call = tool_call.function %}\n                {%- endif %}\n                {{- '<tool_call>\\n{\"name\": \"' }}\n                {{- tool_call.name }}\n                {{- '\", \"arguments\": ' }}\n                {%- if tool_call.arguments is string %}\n                    {{- tool_call.arguments }}\n                {%- else %}\n                    {{- tool_call.arguments | tojson }}\n                {%- endif %}\n                {{- '}\\n</tool_call>' }}\n            {%- endfor %}\n        {%- endif %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if loop.first or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- message.content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n    {%- if enable_thinking is defined and enable_thinking is false %}\n        {{- '<think>\\n\\n</think>\\n\\n' }}\n    {%- endif %}\n{%- endif %}",
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0].role == 'system' %}\n        {{- messages[0].content + '\\n\\n' }}\n    {%- endif %}\n    {{- \"# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n    {%- if messages[0].role == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0].content + '<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}\n{%- for message in messages[::-1] %}\n    {%- set index = (messages|length - 1) - loop.index0 %}\n    {%- if ns.multi_step_tool and message.role == \"user\" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}\n        {%- set ns.multi_step_tool = false %}\n        {%- set ns.last_query_index = index %}\n    {%- endif %}\n{%- endfor %}\n{%- for message in messages %}\n    {%- if message.content is string %}\n        {%- set content = message.content %}\n    {%- else %}\n        {%- set content = '' %}\n    {%- endif %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) %}\n        {{- '<|im_start|>' + message.role + '\\n' + content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {%- set reasoning_content = '' %}\n        {%- if message.reasoning_content is string %}\n            {%- set reasoning_content = message.reasoning_content %}\n        {%- else %}\n            {%- if '</think>' in content %}\n                {%- set reasoning_content = content.split('</think>')[0].rstrip('\\n').split('<think>')[-1].lstrip('\\n') %}\n                {%- set content = content.split('</think>')[-1].lstrip('\\n') %}\n            {%- endif %}\n        {%- endif %}\n        {%- if loop.index0 > ns.last_query_index %}\n            {%- if loop.last or (not loop.last and reasoning_content) %}\n                {{- '<|im_start|>' + message.role + '\\n<think>\\n' + reasoning_content.strip('\\n') + '\\n</think>\\n\\n' + content.lstrip('\\n') }}\n            {%- else %}\n                {{- '<|im_start|>' + message.role + '\\n' + content }}\n            {%- endif %}\n        {%- else %}\n            {{- '<|im_start|>' + message.role + '\\n' + content }}\n        {%- endif %}\n        {%- if message.tool_calls %}\n            {%- for tool_call in message.tool_calls %}\n                {%- if (loop.first and content) or (not loop.first) %}\n                    {{- '\\n' }}\n                {%- endif %}\n                {%- if tool_call.function %}\n                    {%- set tool_call = tool_call.function %}\n                {%- endif %}\n                {{- '<tool_call>\\n{\"name\": \"' }}\n                {{- tool_call.name }}\n                {{- '\", \"arguments\": ' }}\n                {%- if tool_call.arguments is string %}\n                    {{- tool_call.arguments }}\n                {%- else %}\n                    {{- tool_call.arguments | tojson }}\n                {%- endif %}\n                {{- '}\\n</tool_call>' }}\n            {%- endfor %}\n        {%- endif %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if loop.first or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n    {%- if enable_thinking is defined and enable_thinking is false %}\n        {{- '<think>\\n\\n</think>\\n\\n' }}\n    {%- endif %}\n{%- endif %}",
   "clean_up_tokenization_spaces": false,
   "eos_token": "<|im_end|>",
   "errors": "replace",