A Workbench model is a compressed file (in other words, a renamed ZIP file). You should be able to open it with any decent archiver. Inside, there's an XML file. I don't know if the format is documented but it looks quite straightforward:
<value type="object" struct-name="db.mysql.Table" id="{DF899AC1-6CAA-42A8-A584-A663E17596E9}" struct-checksum="0x1a403946">
<value type="string" key="avgRowLength"></value>
<value type="int" key="checksum">0</value>
<value _ptr_="0971BA88" type="list" content-type="object" content-struct-name="db.mysql.Column" key="columns">
<value type="object" struct-name="db.mysql.Column" id="{A742D230-C0CE-48C7-88E0-B5DB74409FE4}" struct-checksum="0x783b5183">
<value type="int" key="autoIncrement">1</value>
<value type="string" key="characterSetName"></value>
<value _ptr_="0971BF08" type="list" content-type="object" content-struct-name="db.CheckConstraint" key="checks"/>
<value type="string" key="collationName"></value>
<value type="string" key="datatypeExplicitParams"></value>
<value type="string" key="defaultValue"></value>
<value type="int" key="defaultValueIsNull">0</value>
<value _ptr_="0971BF98" type="list" content-type="string" key="flags">
<value type="string">UNSIGNED</value>
</value>
<value type="int" key="isNotNull">1</value>
<value type="int" key="length">-1</value>
<value type="string" key="name">invoice_id</value>
<value type="int" key="precision">10</value>
<value type="int" key="scale">-1</value>
<link type="object" struct-name="db.SimpleDatatype" key="simpleType">com.mysql.rdbms.mysql.datatype.int</link>
<value type="string" key="comment"></value>
<value type="string" key="oldName">invoice_id</value>
<link type="object" struct-name="GrtObject" key="owner">{DF899AC1-6CAA-42A8-A584-A663E17596E9}</link>
So you basically need to spend a few days writing a script that parses XML and composes SQL code.
(Honestly, I think it'd be a waste of time, considering that the original Workbench is free and cross-platform.)