Phi-3.5-vision-instruct-fp16-ov / openvino_vision_projection_model.xml
katuni4ka's picture
add model
f0b5cef verified
<?xml version="1.0"?>
<net name="Model3" version="11">
<layers>
<layer id="0" name="input" type="Parameter" version="opset1">
<data shape="?,?,4096" element_type="f32" />
<output>
<port id="0" precision="FP32" names="input">
<dim>-1</dim>
<dim>-1</dim>
<dim>4096</dim>
</port>
</output>
</layer>
<layer id="1" name="self.0.weight_compressed" type="Const" version="opset1">
<data element_type="f16" shape="3072, 4096" offset="0" size="25165824" />
<output>
<port id="0" precision="FP16" names="self.0.weight">
<dim>3072</dim>
<dim>4096</dim>
</port>
</output>
</layer>
<layer id="2" name="self.0.weight" type="Convert" version="opset1">
<data destination_type="f32" />
<rt_info>
<attribute name="decompression" version="0" />
</rt_info>
<input>
<port id="0" precision="FP16">
<dim>3072</dim>
<dim>4096</dim>
</port>
</input>
<output>
<port id="1" precision="FP32">
<dim>3072</dim>
<dim>4096</dim>
</port>
</output>
</layer>
<layer id="3" name="__module.0/aten::linear/MatMul" type="MatMul" version="opset1">
<data transpose_a="false" transpose_b="true" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>4096</dim>
</port>
<port id="1" precision="FP32">
<dim>3072</dim>
<dim>4096</dim>
</port>
</input>
<output>
<port id="2" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>3072</dim>
</port>
</output>
</layer>
<layer id="4" name="Constant_113747_compressed" type="Const" version="opset1">
<data element_type="f16" shape="1, 1, 3072" offset="25165824" size="6144" />
<output>
<port id="0" precision="FP16">
<dim>1</dim>
<dim>1</dim>
<dim>3072</dim>
</port>
</output>
</layer>
<layer id="5" name="Constant_113747" type="Convert" version="opset1">
<data destination_type="f32" />
<rt_info>
<attribute name="decompression" version="0" />
</rt_info>
<input>
<port id="0" precision="FP16">
<dim>1</dim>
<dim>1</dim>
<dim>3072</dim>
</port>
</input>
<output>
<port id="1" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>3072</dim>
</port>
</output>
</layer>
<layer id="6" name="__module.0/aten::linear/Add" type="Add" version="opset1">
<data auto_broadcast="numpy" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>3072</dim>
</port>
<port id="1" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>3072</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="11">
<dim>-1</dim>
<dim>-1</dim>
<dim>3072</dim>
</port>
</output>
</layer>
<layer id="7" name="__module.1/aten::gelu/Gelu" type="Gelu" version="opset7">
<data approximation_mode="ERF" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>3072</dim>
</port>
</input>
<output>
<port id="1" precision="FP32" names="13">
<dim>-1</dim>
<dim>-1</dim>
<dim>3072</dim>
</port>
</output>
</layer>
<layer id="8" name="self.2.weight_compressed" type="Const" version="opset1">
<data element_type="f16" shape="3072, 3072" offset="25171968" size="18874368" />
<output>
<port id="0" precision="FP16" names="self.2.weight">
<dim>3072</dim>
<dim>3072</dim>
</port>
</output>
</layer>
<layer id="9" name="self.2.weight" type="Convert" version="opset1">
<data destination_type="f32" />
<rt_info>
<attribute name="decompression" version="0" />
</rt_info>
<input>
<port id="0" precision="FP16">
<dim>3072</dim>
<dim>3072</dim>
</port>
</input>
<output>
<port id="1" precision="FP32">
<dim>3072</dim>
<dim>3072</dim>
</port>
</output>
</layer>
<layer id="10" name="__module.2/aten::linear/MatMul" type="MatMul" version="opset1">
<data transpose_a="false" transpose_b="true" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>3072</dim>
</port>
<port id="1" precision="FP32">
<dim>3072</dim>
<dim>3072</dim>
</port>
</input>
<output>
<port id="2" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>3072</dim>
</port>
</output>
</layer>
<layer id="11" name="Constant_113748_compressed" type="Const" version="opset1">
<data element_type="f16" shape="1, 1, 3072" offset="44046336" size="6144" />
<output>
<port id="0" precision="FP16">
<dim>1</dim>
<dim>1</dim>
<dim>3072</dim>
</port>
</output>
</layer>
<layer id="12" name="Constant_113748" type="Convert" version="opset1">
<data destination_type="f32" />
<rt_info>
<attribute name="decompression" version="0" />
</rt_info>
<input>
<port id="0" precision="FP16">
<dim>1</dim>
<dim>1</dim>
<dim>3072</dim>
</port>
</input>
<output>
<port id="1" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>3072</dim>
</port>
</output>
</layer>
<layer id="13" name="__module.2/aten::linear/Add" type="Add" version="opset1">
<data auto_broadcast="numpy" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>3072</dim>
</port>
<port id="1" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>3072</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="last_hidden_state">
<dim>-1</dim>
<dim>-1</dim>
<dim>3072</dim>
</port>
</output>
</layer>
<layer id="14" name="Result_112107" type="Result" version="opset1">
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>3072</dim>
</port>
</input>
</layer>
</layers>
<edges>
<edge from-layer="0" from-port="0" to-layer="3" to-port="0" />
<edge from-layer="1" from-port="0" to-layer="2" to-port="0" />
<edge from-layer="2" from-port="1" to-layer="3" to-port="1" />
<edge from-layer="3" from-port="2" to-layer="6" to-port="0" />
<edge from-layer="4" from-port="0" to-layer="5" to-port="0" />
<edge from-layer="5" from-port="1" to-layer="6" to-port="1" />
<edge from-layer="6" from-port="2" to-layer="7" to-port="0" />
<edge from-layer="7" from-port="1" to-layer="10" to-port="0" />
<edge from-layer="8" from-port="0" to-layer="9" to-port="0" />
<edge from-layer="9" from-port="1" to-layer="10" to-port="1" />
<edge from-layer="10" from-port="2" to-layer="13" to-port="0" />
<edge from-layer="11" from-port="0" to-layer="12" to-port="0" />
<edge from-layer="12" from-port="1" to-layer="13" to-port="1" />
<edge from-layer="13" from-port="2" to-layer="14" to-port="0" />
</edges>
<rt_info>
<Runtime_version value="2025.0.0-17908-513dcc5c7b7-releases/2025/0" />
<conversion_parameters>
<framework value="pytorch" />
<is_python_object value="True" />
</conversion_parameters>
<optimum>
<optimum_intel_version value="1.22.0.dev0+e465c7f7" />
<optimum_version value="1.24.0.dev0" />
<pytorch_version value="2.5.1+cpu" />
<transformers_version value="4.47.0" />
</optimum>
</rt_info>
</net>