Upload 12 files

.gitattributes

@@ -35,3 +35,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
 Mistral-NeMo-Minitron-8B-Instruct-int4_asym-awq-se-ov/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 Mistral-NeMo-Minitron-8B-Instruct-int4_asym-ov/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Mistral-NeMo-Minitron-8B-Instruct-int4_asym-gptq-ov/tokenizer.json filter=lfs diff=lfs merge=lfs -text

Mistral-NeMo-Minitron-8B-Instruct-int4_asym-gptq-ov/config.json

@@ -0,0 +1,27 @@
+{
+  "activation": "silu",
+  "architectures": [
+    "MistralForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 11520,
+  "max_position_embeddings": 8192,
+  "model_type": "mistral",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 40,
+  "num_key_value_heads": 8,
+  "rms_norm_eps": 1e-05,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.51.3",
+  "use_cache": true,
+  "vocab_size": 131072
+}

Mistral-NeMo-Minitron-8B-Instruct-int4_asym-gptq-ov/generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "transformers_version": "4.51.3"
+}

Mistral-NeMo-Minitron-8B-Instruct-int4_asym-gptq-ov/openvino_config.json

@@ -0,0 +1,27 @@
+{
+  "dtype": "int4",
+  "input_info": null,
+  "optimum_version": "1.25.2",
+  "quantization_config": {
+    "all_layers": null,
+    "backup_precision": null,
+    "bits": 4,
+    "dataset": "wikitext2",
+    "dtype": "int4",
+    "gptq": true,
+    "group_size": 128,
+    "ignored_scope": null,
+    "lora_correction": null,
+    "num_samples": null,
+    "processor": null,
+    "quant_method": "default",
+    "ratio": 0.8,
+    "scale_estimation": null,
+    "sensitivity_metric": null,
+    "sym": false,
+    "tokenizer": null,
+    "trust_remote_code": false
+  },
+  "save_onnx_model": false,
+  "transformers_version": "4.51.3"
+}

Mistral-NeMo-Minitron-8B-Instruct-int4_asym-gptq-ov/openvino_detokenizer.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e3f29ca8f5d47f3ff16501cfbfa42f1e39a8600f6d0e9f85b7890835d53e3e9b
+size 1943696

Mistral-NeMo-Minitron-8B-Instruct-int4_asym-gptq-ov/openvino_detokenizer.xml

@@ -0,0 +1,220 @@
+<?xml version="1.0"?>
+<net name="detokenizer" version="11">
+	<layers>
+		<layer id="0" name="Parameter_2894600" type="Parameter" version="opset1">
+			<data shape="?,?" element_type="i64" />
+			<output>
+				<port id="0" precision="I64" names="Parameter_2894600">
+					<dim>-1</dim>
+					<dim>-1</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="1" name="Convert_2894770" type="Convert" version="opset1">
+			<data destination_type="i32" />
+			<input>
+				<port id="0" precision="I64">
+					<dim>-1</dim>
+					<dim>-1</dim>
+				</port>
+			</input>
+			<output>
+				<port id="1" precision="I32">
+					<dim>-1</dim>
+					<dim>-1</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="2" name="Constant_2894602" type="Const" version="opset1">
+			<data element_type="i32" shape="131072" offset="0" size="524288" />
+			<output>
+				<port id="0" precision="I32">
+					<dim>131072</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="3" name="Constant_2894604" type="Const" version="opset1">
+			<data element_type="i32" shape="131072" offset="524288" size="524288" />
+			<output>
+				<port id="0" precision="I32">
+					<dim>131072</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="4" name="Constant_2894606" type="Const" version="opset1">
+			<data element_type="u8" shape="891120" offset="1048576" size="891120" />
+			<output>
+				<port id="0" precision="U8">
+					<dim>891120</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="5" name="Slice_2894611" type="Const" version="opset1">
+			<data element_type="i32" shape="1000" offset="1939696" size="4000" />
+			<output>
+				<port id="0" precision="I32">
+					<dim>1000</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="6" name="VocabDecoder_2894613" type="VocabDecoder" version="extension">
+			<data skip_tokens="" />
+			<input>
+				<port id="0" precision="I32">
+					<dim>-1</dim>
+					<dim>-1</dim>
+				</port>
+				<port id="1" precision="I32">
+					<dim>131072</dim>
+				</port>
+				<port id="2" precision="I32">
+					<dim>131072</dim>
+				</port>
+				<port id="3" precision="U8">
+					<dim>891120</dim>
+				</port>
+				<port id="4" precision="I32">
+					<dim>1000</dim>
+				</port>
+			</input>
+			<output>
+				<port id="5" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<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>
+				<port id="9" precision="U8">
+					<dim>-1</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="7" name="FuzeRagged_2894614" type="FuzeRagged" version="extension">
+			<input>
+				<port id="0" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="1" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="2" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="3" precision="I32">
+					<dim>-1</dim>
+				</port>
+			</input>
+			<output>
+				<port id="4" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="5" precision="I32">
+					<dim>-1</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="8" name="UTF8Validate_2894615" type="UTF8Validate" version="extension">
+			<data replace_mode="true" />
+			<input>
+				<port id="0" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="1" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="2" precision="U8">
+					<dim>-1</dim>
+				</port>
+			</input>
+			<output>
+				<port id="3" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="4" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="5" precision="U8">
+					<dim>-1</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="9" name="StringTensorPack_2894616" type="StringTensorPack" version="opset15">
+			<input>
+				<port id="0" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="1" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="2" precision="U8">
+					<dim>-1</dim>
+				</port>
+			</input>
+			<output>
+				<port id="3" precision="STRING" names="Result_2894617,string_output">
+					<dim>-1</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="10" name="Result_2894617" type="Result" version="opset1" output_names="Result_2894617,string_output">
+			<input>
+				<port id="0" precision="STRING">
+					<dim>-1</dim>
+				</port>
+			</input>
+		</layer>
+	</layers>
+	<edges>
+		<edge from-layer="0" from-port="0" to-layer="1" to-port="0" />
+		<edge from-layer="1" from-port="1" to-layer="6" to-port="0" />
+		<edge from-layer="2" from-port="0" to-layer="6" to-port="1" />
+		<edge from-layer="3" from-port="0" to-layer="6" to-port="2" />
+		<edge from-layer="4" from-port="0" to-layer="6" to-port="3" />
+		<edge from-layer="5" from-port="0" to-layer="6" to-port="4" />
+		<edge from-layer="6" from-port="7" to-layer="7" to-port="2" />
+		<edge from-layer="6" from-port="9" to-layer="8" to-port="2" />
+		<edge from-layer="6" from-port="8" to-layer="7" to-port="3" />
+		<edge from-layer="6" from-port="6" to-layer="7" to-port="1" />
+		<edge from-layer="6" from-port="5" to-layer="7" to-port="0" />
+		<edge from-layer="7" from-port="4" to-layer="8" to-port="0" />
+		<edge from-layer="7" from-port="5" to-layer="8" to-port="1" />
+		<edge from-layer="8" from-port="3" to-layer="9" to-port="0" />
+		<edge from-layer="8" from-port="4" to-layer="9" to-port="1" />
+		<edge from-layer="8" from-port="5" to-layer="9" to-port="2" />
+		<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="1" />
+		<chat_template value="{{'&lt;extra_id_0>System'}}{% for message in messages %}{% if message['role'] == 'system' %}{{'&#10;' + message['content'].strip()}}{% endif %}{% endfor %}{{'&#10;'}}{% for message in messages %}{% if message['role'] == 'user' %}{{ '&#10;&lt;extra_id_1>User&#10;' + message['content'].strip() + '&#10;&lt;extra_id_1>Assistant&#10;' }}{% elif message['role'] == 'assistant' %}{{ message['content'].strip() }}{% endif %}{% endfor %}" />
+		<clean_up_tokenization_spaces />
+		<detokenizer_input_type value="i64" />
+		<eos_token_id value="2" />
+		<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" />
+		<original_post_processor_template value="{&quot;type&quot;: &quot;TemplateProcessing&quot;, &quot;single&quot;: [{&quot;SpecialToken&quot;: {&quot;id&quot;: &quot;&lt;s>&quot;, &quot;type_id&quot;: 0}}, {&quot;Sequence&quot;: {&quot;id&quot;: &quot;A&quot;, &quot;type_id&quot;: 0}}], &quot;pair&quot;: [{&quot;SpecialToken&quot;: {&quot;id&quot;: &quot;&lt;s>&quot;, &quot;type_id&quot;: 0}}, {&quot;Sequence&quot;: {&quot;id&quot;: &quot;A&quot;, &quot;type_id&quot;: 0}}, {&quot;SpecialToken&quot;: {&quot;id&quot;: &quot;&lt;s>&quot;, &quot;type_id&quot;: 1}}, {&quot;Sequence&quot;: {&quot;id&quot;: &quot;B&quot;, &quot;type_id&quot;: 1}}], &quot;special_tokens&quot;: {&quot;&lt;/s>&quot;: {&quot;id&quot;: &quot;&lt;/s>&quot;, &quot;ids&quot;: [2], &quot;tokens&quot;: [&quot;&lt;/s>&quot;]}, &quot;&lt;s>&quot;: {&quot;id&quot;: &quot;&lt;s>&quot;, &quot;ids&quot;: [1], &quot;tokens&quot;: [&quot;&lt;s>&quot;]}}}" />
+		<original_tokenizer_class value="&lt;class 'transformers.tokenization_utils_fast.PreTrainedTokenizerFast'>" />
+		<processed_post_processor_template value="{&quot;single&quot;: {&quot;ids&quot;: [1, -1], &quot;type_ids&quot;: [0, 0]}, &quot;pair&quot;: {&quot;ids&quot;: [1, -1, 1, -2], &quot;type_ids&quot;: [0, 0, 1, 1]}}" />
+		<sentencepiece_version value="0.2.0" />
+		<skip_special_tokens value="True" />
+		<streaming_detokenizer value="False" />
+		<tokenizer_output_type value="i64" />
+		<tokenizers_version value="0.21.1" />
+		<transformers_version value="4.51.3" />
+		<use_max_padding value="False" />
+		<use_sentencepiece_backend value="False" />
+		<utf8_replace_mode value="replace" />
+		<with_detokenizer value="True" />
+	</rt_info>
+</net>

Mistral-NeMo-Minitron-8B-Instruct-int4_asym-gptq-ov/openvino_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f38cb9e5ee0242b49228f061f05843fc7310b00b18e151d2d53420128482d82b
+size 5596009292

Mistral-NeMo-Minitron-8B-Instruct-int4_asym-gptq-ov/openvino_tokenizer.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c0b12029962fe8e29f8c9a385c32bf5af44731d897ed8beef1d5d1e66896b988
+size 8151102

Mistral-NeMo-Minitron-8B-Instruct-int4_asym-gptq-ov/openvino_tokenizer.xml

@@ -0,0 +1,762 @@
+<?xml version="1.0"?>
+<net name="tokenizer" version="11">
+	<layers>
+		<layer id="0" name="Parameter_2894474" type="Parameter" version="opset1">
+			<data shape="?" element_type="string" />
+			<output>
+				<port id="0" precision="STRING" names="Parameter_2894474">
+					<dim>-1</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="1" name="Constant_2894584" 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_2894585" 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_2894586" type="Const" version="opset1">
+			<data element_type="i32" shape="1" offset="4" size="4" />
+			<output>
+				<port id="0" precision="I32">
+					<dim>1</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="4" name="Constant_2894480" type="Const" version="opset1">
+			<data element_type="i64" shape="" offset="8" size="8" />
+			<output>
+				<port id="0" precision="I64" />
+			</output>
+		</layer>
+		<layer id="5" name="StringTensorUnpack_2894475" type="StringTensorUnpack" version="opset15">
+			<input>
+				<port id="0" precision="STRING">
+					<dim>-1</dim>
+				</port>
+			</input>
+			<output>
+				<port id="1" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="2" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="3" precision="U8">
+					<dim>-1</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="6" name="ShapeOf_2894476" type="ShapeOf" version="opset3">
+			<data output_type="i64" />
+			<input>
+				<port id="0" precision="I32">
+					<dim>-1</dim>
+				</port>
+			</input>
+			<output>
+				<port id="1" precision="I64">
+					<dim>1</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="7" name="Constant_2894477" 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="Constant_2894478" type="Const" version="opset1">
+			<data element_type="i64" shape="" offset="8" size="8" />
+			<output>
+				<port id="0" precision="I64" />
+			</output>
+		</layer>
+		<layer id="9" name="Gather_2894479" type="Gather" version="opset8">
+			<data batch_dims="0" />
+			<input>
+				<port id="0" precision="I64">
+					<dim>1</dim>
+				</port>
+				<port id="1" precision="I64" />
+				<port id="2" precision="I64" />
+			</input>
+			<output>
+				<port id="3" precision="I64" />
+			</output>
+		</layer>
+		<layer id="10" name="Constant_2894481" 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="Range_2894482" type="Range" version="opset4">
+			<data output_type="i32" />
+			<input>
+				<port id="0" precision="I64" />
+				<port id="1" precision="I64" />
+				<port id="2" precision="I64" />
+			</input>
+			<output>
+				<port id="3" precision="I32">
+					<dim>-1</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="12" name="Constant_2894483" type="Const" version="opset1">
+			<data element_type="i64" shape="" offset="16" size="8" />
+			<output>
+				<port id="0" precision="I64" />
+			</output>
+		</layer>
+		<layer id="13" name="Constant_2894484" type="Const" version="opset1">
+			<data element_type="i64" shape="" offset="16" size="8" />
+			<output>
+				<port id="0" precision="I64" />
+			</output>
+		</layer>
+		<layer id="14" name="Add_2894485" type="Add" version="opset1">
+			<data auto_broadcast="numpy" />
+			<input>
+				<port id="0" precision="I64" />
+				<port id="1" precision="I64" />
+			</input>
+			<output>
+				<port id="2" precision="I64" />
+			</output>
+		</layer>
+		<layer id="15" name="Constant_2894486" type="Const" version="opset1">
+			<data element_type="i64" shape="" offset="16" size="8" />
+			<output>
+				<port id="0" precision="I64" />
+			</output>
+		</layer>
+		<layer id="16" name="Range_2894487" type="Range" version="opset4">
+			<data output_type="i32" />
+			<input>
+				<port id="0" precision="I64" />
+				<port id="1" precision="I64" />
+				<port id="2" precision="I64" />
+			</input>
+			<output>
+				<port id="3" precision="I32">
+					<dim>-1</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="17" name="Constant_2894549" type="Const" version="opset1">
+			<data element_type="u8" shape="17865" offset="24" size="17865" />
+			<output>
+				<port id="0" precision="U8">
+					<dim>17865</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="18" name="SpecialTokensSplit_2894550" type="SpecialTokensSplit" version="extension">
+			<input>
+				<port id="0" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="1" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="2" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="3" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="4" precision="U8">
+					<dim>-1</dim>
+				</port>
+				<port id="5" precision="U8">
+					<dim>17865</dim>
+				</port>
+			</input>
+			<output>
+				<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>
+				<port id="9" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="10" precision="U8">
+					<dim>-1</dim>
+				</port>
+				<port id="11" precision="BOOL">
+					<dim>-1</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="19" name="Constant_2894552" type="Const" version="opset1">
+			<data element_type="u8" shape="211" offset="17889" size="211" />
+			<output>
+				<port id="0" precision="U8">
+					<dim>211</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="20" name="RegexSplit_2894553" type="RegexSplit" version="extension">
+			<data behaviour="isolate" invert="false" max_splits="-1" />
+			<input>
+				<port id="0" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="1" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="2" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="3" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="4" precision="U8">
+					<dim>-1</dim>
+				</port>
+				<port id="5" precision="BOOL">
+					<dim>-1</dim>
+				</port>
+				<port id="6" precision="U8">
+					<dim>211</dim>
+				</port>
+			</input>
+			<output>
+				<port id="7" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="8" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="9" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="10" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="11" precision="U8">
+					<dim>-1</dim>
+				</port>
+				<port id="12" precision="BOOL">
+					<dim>-1</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="21" name="Constant_2894555" type="Const" version="opset1">
+			<data element_type="i32" shape="131072" offset="18100" size="524288" />
+			<output>
+				<port id="0" precision="I32">
+					<dim>131072</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="22" name="Constant_2894557" type="Const" version="opset1">
+			<data element_type="i32" shape="131072" offset="542388" size="524288" />
+			<output>
+				<port id="0" precision="I32">
+					<dim>131072</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="23" name="Constant_2894559" type="Const" version="opset1">
+			<data element_type="u8" shape="891120" offset="1066676" size="891120" />
+			<output>
+				<port id="0" precision="U8">
+					<dim>891120</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="24" name="Constant_2894567" type="Const" version="opset1">
+			<data element_type="i32" shape="269443" offset="1957796" size="1077772" />
+			<output>
+				<port id="0" precision="I32">
+					<dim>269443</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="25" name="Constant_2894569" type="Const" version="opset1">
+			<data element_type="i32" shape="269443" offset="3035568" size="1077772" />
+			<output>
+				<port id="0" precision="I32">
+					<dim>269443</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="26" name="Constant_2894571" type="Const" version="opset1">
+			<data element_type="u8" shape="989624" offset="4113340" size="989624" />
+			<output>
+				<port id="0" precision="U8">
+					<dim>989624</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="27" name="Constant_2894573" type="Const" version="opset1">
+			<data element_type="i32" shape="269443" offset="5102964" size="1077772" />
+			<output>
+				<port id="0" precision="I32">
+					<dim>269443</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="28" name="Constant_2894575" type="Const" version="opset1">
+			<data element_type="i32" shape="269443" offset="6180736" size="1077772" />
+			<output>
+				<port id="0" precision="I32">
+					<dim>269443</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="29" name="Constant_2894577" type="Const" version="opset1">
+			<data element_type="u8" shape="867745" offset="7258508" size="867745" />
+			<output>
+				<port id="0" precision="U8">
+					<dim>867745</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="30" name="Constant_2894561" type="Const" version="opset1">
+			<data element_type="i32" shape="999" offset="8126253" size="3996" />
+			<output>
+				<port id="0" precision="I32">
+					<dim>999</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="31" name="Constant_2894563" type="Const" version="opset1">
+			<data element_type="i32" shape="999" offset="8130249" size="3996" />
+			<output>
+				<port id="0" precision="I32">
+					<dim>999</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="32" name="Constant_2894565" type="Const" version="opset1">
+			<data element_type="u8" shape="12857" offset="8134245" size="12857" />
+			<output>
+				<port id="0" precision="U8">
+					<dim>12857</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="33" name="Constant_2894578" type="Const" version="opset1">
+			<data element_type="i32" shape="999" offset="8147102" size="3996" />
+			<output>
+				<port id="0" precision="I32">
+					<dim>999</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="34" name="BPETokenizer_2894579" type="BPETokenizer" version="extension">
+			<data unk_token="" fuse_unk="false" suffix_indicator="" end_suffix="" byte_fallback="false" cache_capacity="26214" />
+			<input>
+				<port id="0" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="1" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="2" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="3" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="4" precision="U8">
+					<dim>-1</dim>
+				</port>
+				<port id="5" precision="I32">
+					<dim>131072</dim>
+				</port>
+				<port id="6" precision="I32">
+					<dim>131072</dim>
+				</port>
+				<port id="7" precision="U8">
+					<dim>891120</dim>
+				</port>
+				<port id="8" precision="I32">
+					<dim>269443</dim>
+				</port>
+				<port id="9" precision="I32">
+					<dim>269443</dim>
+				</port>
+				<port id="10" precision="U8">
+					<dim>989624</dim>
+				</port>
+				<port id="11" precision="I32">
+					<dim>269443</dim>
+				</port>
+				<port id="12" precision="I32">
+					<dim>269443</dim>
+				</port>
+				<port id="13" precision="U8">
+					<dim>867745</dim>
+				</port>
+				<port id="14" precision="I32">
+					<dim>999</dim>
+				</port>
+				<port id="15" precision="I32">
+					<dim>999</dim>
+				</port>
+				<port id="16" precision="U8">
+					<dim>12857</dim>
+				</port>
+				<port id="17" precision="I32">
+					<dim>999</dim>
+				</port>
+			</input>
+			<output>
+				<port id="18" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="19" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="20" precision="I32">
+					<dim>-1</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="35" name="Subtract_2894580" 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="36" name="Constant_2894581" type="Const" version="opset1">
+			<data element_type="i32" shape="" offset="8151098" size="4" />
+			<output>
+				<port id="0" precision="I32" />
+			</output>
+		</layer>
+		<layer id="37" name="Minimum_2894582" type="Minimum" version="opset1">
+			<data auto_broadcast="numpy" />
+			<input>
+				<port id="0" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="1" precision="I32" />
+			</input>
+			<output>
+				<port id="2" precision="I32">
+					<dim>-1</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="38" name="Subtract_2894583" 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="39" name="Constant_2894587" type="Const" version="opset1">
+			<data element_type="i32" shape="2" offset="8" size="8" />
+			<output>
+				<port id="0" precision="I32">
+					<dim>2</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="40" name="CombineSegments_2894588" type="CombineSegments" version="extension">
+			<input>
+				<port id="0" precision="I32" />
+				<port id="1" precision="I32" />
+				<port id="2" precision="I32">
+					<dim>1</dim>
+				</port>
+				<port id="3" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="4" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="5" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="6" precision="I32">
+					<dim>2</dim>
+				</port>
+			</input>
+			<output>
+				<port id="7" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="8" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="9" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="10" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="11" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="12" precision="I32">
+					<dim>-1</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="41" name="Subtract_2894589" 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="42" name="Constant_2894590" type="Const" version="opset1">
+			<data element_type="i32" shape="" offset="0" size="4" />
+			<output>
+				<port id="0" precision="I32" />
+			</output>
+		</layer>
+		<layer id="43" name="ReduceMax_2894591" type="ReduceMax" version="opset1">
+			<data keep_dims="false" />
+			<input>
+				<port id="0" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<port id="1" precision="I32" />
+			</input>
+			<output>
+				<port id="2" precision="I32" />
+			</output>
+		</layer>
+		<layer id="44" name="Constant_2894592" 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="RaggedToDense_2894593" type="RaggedToDense" version="extension">
+			<data pad_right="false" m_pad_max_length="false" />
+			<input>
+				<port id="0" precision="I32">
+					<dim>-1</dim>
+				</port>
+				<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="I32" />
+			</input>
+			<output>
+				<port id="5" precision="I32">
+					<dim>-1</dim>
+					<dim>-1</dim>
+				</port>
+				<port id="6" precision="BOOL">
+					<dim>-1</dim>
+					<dim>-1</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="46" name="Convert_2894594" type="Convert" version="opset1">
+			<data destination_type="i32" />
+			<input>
+				<port id="0" precision="BOOL">
+					<dim>-1</dim>
+					<dim>-1</dim>
+				</port>
+			</input>
+			<output>
+				<port id="1" precision="I32">
+					<dim>-1</dim>
+					<dim>-1</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="47" name="Convert_2894594.0" type="Convert" version="opset1">
+			<data destination_type="i64" />
+			<input>
+				<port id="0" precision="I32">
+					<dim>-1</dim>
+					<dim>-1</dim>
+				</port>
+			</input>
+			<output>
+				<port id="1" precision="I64" names="attention_mask">
+					<dim>-1</dim>
+					<dim>-1</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="49" name="RaggedToDense_2894593.0" type="Convert" version="opset1">
+			<data destination_type="i64" />
+			<input>
+				<port id="0" precision="I32">
+					<dim>-1</dim>
+					<dim>-1</dim>
+				</port>
+			</input>
+			<output>
+				<port id="1" precision="I64" names="input_ids">
+					<dim>-1</dim>
+					<dim>-1</dim>
+				</port>
+			</output>
+		</layer>
+		<layer id="50" name="Result_2894597" type="Result" version="opset1" output_names="input_ids">
+			<input>
+				<port id="0" precision="I64">
+					<dim>-1</dim>
+					<dim>-1</dim>
+				</port>
+			</input>
+		</layer>
+		<layer id="48" name="Result_2894599" type="Result" version="opset1" output_names="attention_mask">
+			<input>
+				<port id="0" precision="I64">
+					<dim>-1</dim>
+					<dim>-1</dim>
+				</port>
+			</input>
+		</layer>
+	</layers>
+	<edges>
+		<edge from-layer="0" from-port="0" to-layer="5" to-port="0" />
+		<edge from-layer="1" from-port="0" to-layer="40" to-port="0" />
+		<edge from-layer="2" from-port="0" to-layer="40" to-port="1" />
+		<edge from-layer="3" from-port="0" to-layer="40" to-port="2" />
+		<edge from-layer="4" from-port="0" to-layer="11" to-port="0" />
+		<edge from-layer="5" from-port="1" to-layer="6" to-port="0" />
+		<edge from-layer="5" from-port="3" to-layer="18" to-port="4" />
+		<edge from-layer="5" from-port="2" to-layer="18" to-port="3" />
+		<edge from-layer="5" from-port="1" to-layer="18" to-port="2" />
+		<edge from-layer="6" from-port="1" to-layer="9" to-port="0" />
+		<edge from-layer="7" from-port="0" to-layer="9" to-port="1" />
+		<edge from-layer="8" from-port="0" to-layer="9" to-port="2" />
+		<edge from-layer="9" from-port="3" to-layer="14" to-port="0" />
+		<edge from-layer="9" from-port="3" to-layer="11" to-port="1" />
+		<edge from-layer="10" from-port="0" to-layer="11" to-port="2" />
+		<edge from-layer="11" from-port="3" to-layer="18" to-port="0" />
+		<edge from-layer="12" from-port="0" to-layer="16" to-port="0" />
+		<edge from-layer="13" from-port="0" to-layer="14" to-port="1" />
+		<edge from-layer="14" from-port="2" to-layer="16" to-port="1" />
+		<edge from-layer="15" from-port="0" to-layer="16" to-port="2" />
+		<edge from-layer="16" from-port="3" to-layer="18" to-port="1" />
+		<edge from-layer="17" from-port="0" to-layer="18" to-port="5" />
+		<edge from-layer="18" from-port="11" to-layer="20" to-port="5" />
+		<edge from-layer="18" from-port="10" to-layer="20" to-port="4" />
+		<edge from-layer="18" from-port="9" to-layer="20" to-port="3" />
+		<edge from-layer="18" from-port="8" to-layer="20" to-port="2" />
+		<edge from-layer="18" from-port="7" to-layer="20" to-port="1" />
+		<edge from-layer="18" from-port="6" to-layer="20" to-port="0" />
+		<edge from-layer="19" from-port="0" to-layer="20" to-port="6" />
+		<edge from-layer="20" from-port="7" to-layer="34" to-port="0" />
+		<edge from-layer="20" from-port="8" to-layer="34" to-port="1" />
+		<edge from-layer="20" from-port="9" to-layer="34" to-port="2" />
+		<edge from-layer="20" from-port="10" to-layer="34" to-port="3" />
+		<edge from-layer="20" from-port="11" to-layer="34" to-port="4" />
+		<edge from-layer="21" from-port="0" to-layer="34" to-port="5" />
+		<edge from-layer="22" from-port="0" to-layer="34" to-port="6" />
+		<edge from-layer="23" from-port="0" to-layer="34" to-port="7" />
+		<edge from-layer="24" from-port="0" to-layer="34" to-port="8" />
+		<edge from-layer="25" from-port="0" to-layer="34" to-port="9" />
+		<edge from-layer="26" from-port="0" to-layer="34" to-port="10" />
+		<edge from-layer="27" from-port="0" to-layer="34" to-port="11" />
+		<edge from-layer="28" from-port="0" to-layer="34" to-port="12" />
+		<edge from-layer="29" from-port="0" to-layer="34" to-port="13" />
+		<edge from-layer="30" from-port="0" to-layer="34" to-port="14" />
+		<edge from-layer="31" from-port="0" to-layer="34" to-port="15" />
+		<edge from-layer="32" from-port="0" to-layer="34" to-port="16" />
+		<edge from-layer="33" from-port="0" to-layer="34" to-port="17" />
+		<edge from-layer="34" from-port="20" to-layer="40" to-port="5" />
+		<edge from-layer="34" from-port="19" to-layer="40" to-port="4" />
+		<edge from-layer="34" from-port="19" to-layer="38" to-port="0" />
+		<edge from-layer="34" from-port="18" to-layer="35" to-port="1" />
+		<edge from-layer="34" from-port="19" to-layer="35" to-port="0" />
+		<edge from-layer="35" from-port="2" to-layer="37" to-port="0" />
+		<edge from-layer="36" from-port="0" to-layer="37" to-port="1" />
+		<edge from-layer="37" from-port="2" to-layer="38" to-port="1" />
+		<edge from-layer="38" from-port="2" to-layer="40" to-port="3" />
+		<edge from-layer="39" from-port="0" to-layer="40" to-port="6" />
+		<edge from-layer="40" from-port="8" to-layer="41" to-port="0" />
+		<edge from-layer="40" from-port="7" to-layer="41" to-port="1" />
+		<edge from-layer="40" from-port="7" to-layer="45" to-port="0" />
+		<edge from-layer="40" from-port="8" to-layer="45" to-port="1" />
+		<edge from-layer="40" from-port="9" to-layer="45" to-port="2" />
+		<edge from-layer="41" from-port="2" to-layer="43" to-port="0" />
+		<edge from-layer="42" from-port="0" to-layer="43" to-port="1" />
+		<edge from-layer="43" from-port="2" to-layer="45" to-port="3" />
+		<edge from-layer="44" from-port="0" to-layer="45" to-port="4" />
+		<edge from-layer="45" from-port="6" to-layer="46" to-port="0" />
+		<edge from-layer="45" from-port="5" to-layer="49" to-port="0" />
+		<edge from-layer="46" from-port="1" to-layer="47" to-port="0" />
+		<edge from-layer="47" from-port="1" to-layer="48" to-port="0" />
+		<edge from-layer="49" from-port="1" to-layer="50" to-port="0" />
+	</edges>
+	<rt_info>
+		<add_attention_mask value="True" />
+		<add_prefix_space />
+		<add_special_tokens value="True" />
+		<bos_token_id value="1" />
+		<chat_template value="{{'&lt;extra_id_0>System'}}{% for message in messages %}{% if message['role'] == 'system' %}{{'&#10;' + message['content'].strip()}}{% endif %}{% endfor %}{{'&#10;'}}{% for message in messages %}{% if message['role'] == 'user' %}{{ '&#10;&lt;extra_id_1>User&#10;' + message['content'].strip() + '&#10;&lt;extra_id_1>Assistant&#10;' }}{% elif message['role'] == 'assistant' %}{{ message['content'].strip() }}{% endif %}{% endfor %}" />
+		<clean_up_tokenization_spaces />
+		<detokenizer_input_type value="i64" />
+		<eos_token_id value="2" />
+		<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" />
+		<original_post_processor_template value="{&quot;type&quot;: &quot;TemplateProcessing&quot;, &quot;single&quot;: [{&quot;SpecialToken&quot;: {&quot;id&quot;: &quot;&lt;s>&quot;, &quot;type_id&quot;: 0}}, {&quot;Sequence&quot;: {&quot;id&quot;: &quot;A&quot;, &quot;type_id&quot;: 0}}], &quot;pair&quot;: [{&quot;SpecialToken&quot;: {&quot;id&quot;: &quot;&lt;s>&quot;, &quot;type_id&quot;: 0}}, {&quot;Sequence&quot;: {&quot;id&quot;: &quot;A&quot;, &quot;type_id&quot;: 0}}, {&quot;SpecialToken&quot;: {&quot;id&quot;: &quot;&lt;s>&quot;, &quot;type_id&quot;: 1}}, {&quot;Sequence&quot;: {&quot;id&quot;: &quot;B&quot;, &quot;type_id&quot;: 1}}], &quot;special_tokens&quot;: {&quot;&lt;/s>&quot;: {&quot;id&quot;: &quot;&lt;/s>&quot;, &quot;ids&quot;: [2], &quot;tokens&quot;: [&quot;&lt;/s>&quot;]}, &quot;&lt;s>&quot;: {&quot;id&quot;: &quot;&lt;s>&quot;, &quot;ids&quot;: [1], &quot;tokens&quot;: [&quot;&lt;s>&quot;]}}}" />
+		<original_tokenizer_class value="&lt;class 'transformers.tokenization_utils_fast.PreTrainedTokenizerFast'>" />
+		<processed_post_processor_template value="{&quot;single&quot;: {&quot;ids&quot;: [1, -1], &quot;type_ids&quot;: [0, 0]}, &quot;pair&quot;: {&quot;ids&quot;: [1, -1, 1, -2], &quot;type_ids&quot;: [0, 0, 1, 1]}}" />
+		<sentencepiece_version value="0.2.0" />
+		<skip_special_tokens value="True" />
+		<streaming_detokenizer value="False" />
+		<tokenizer_output_type value="i64" />
+		<tokenizers_version value="0.21.1" />
+		<transformers_version value="4.51.3" />
+		<use_max_padding value="False" />
+		<use_sentencepiece_backend value="False" />
+		<utf8_replace_mode value="replace" />
+		<with_detokenizer value="True" />
+	</rt_info>
+</net>

Mistral-NeMo-Minitron-8B-Instruct-int4_asym-gptq-ov/special_tokens_map.json

@@ -0,0 +1,23 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

Mistral-NeMo-Minitron-8B-Instruct-int4_asym-gptq-ov/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b0240ce510f08e6c2041724e9043e33be9d251d1e4a4d94eb68cd47b954b61d2
+size 17078292