using nested classes for organize Model, how to "crossreference"

Question

I'm want to use nested classes to organize my pydantic models. So something like this:

class AttributeValue(BaseModel):
  id: int
  typeid: int
  value: str

class AttributeValueAddRequest(BaseModel):
  typeid: int
  value: str

class AttributeType(BaseModel):
  id: int
  name: str
  type: Literal['a', 'b', 'c']
  values: list[AttributeValue] = None
  extra: bool = False

class AttributeTypeAddRequest(BaseModel):
  name: str
  type: Literal['a', 'b', 'c']
  extra: bool = False

[...]

This is just a small example. But the amount of code grows and it became (in my opinion) very unreadable. So I though, it could be nice to organize this in nested classes:

class MAttribute:
  class Value:
    class Entry(BaseModel):
      id: int
      typeid: int
      value: str

    class AddRequest(BaseModel):
      typeid: int
      value: str


  class Type:
    class Entry(BaseModel):
      id: int
      name: str
      type: Literal['a', 'b', 'c']
      values: list[Value.Entry] = None   <-- How to reference?
      extra: bool = False

    class AddRequest(BaseModel):
      name: str
      type: Literal['a', 'b', 'c']
      extra: bool = False

Attribute=MAttribute()

So if I want to create a Value-Entry model from a dict i use:

  model=Attribute.Value.Entry.parse_obj(**dict)

This works as expected. My problem is now, that I do not know how to reference the values in in Type.Entry model to Value.Entry. I tried it with lambda, etc.. But no luck (or I didn't understand it right).

So how can I do this? Or is there a better way to organize the models and don't have everything in a single class from top to bottom (like in the first code block)?

Regards.