Is there a way to exclude primary key attribute when doing COUNT + LEFT JOIN + GROUP BY + ORDER BY?

I have the following code, its in coffeescript:

Reservation.findAll({
    attributes: [
      [DB.Sequelize.fn('COUNT', 'item_id'), 'num_reservations']
    ]
    include: [
      {model: models.Item, as: 'Item', attributes: []}
    ],
    group: ['Item.id']
    order: 'num_reservations ASC'
    })

that generates the following SQL:

SELECT "reservation"."id", COUNT('item_id') AS "num_reservations", "Item"."id" AS "Item.id" 
FROM "reservation" AS "reservation" 
LEFT OUTER JOIN "item" AS "Item" ON "reservation"."item_id" = "Item"."id" AND "Item"."deleted_at" IS NULL 
WHERE "reservation"."deleted_at" IS NULL GROUP BY "Item"."id" ORDER BY num_reservations ASC

and the error:

SequelizeDatabaseError: column "res_reservation.id" must appear in the GROUP BY clause

The expected SQL that works would be the same as above just without "reservation"."id" from the SELECT statement:

SELECT COUNT('item_id') AS "num_reservations", "Item"."id" AS "Item.id" 
FROM "reservation" AS "reservation" 
LEFT OUTER JOIN "item" AS "Item" ON "reservation"."item_id" = "Item"."id" AND "Item"."deleted_at" IS NULL 
WHERE "reservation"."deleted_at" IS NULL GROUP BY "Item"."id" ORDER BY num_reservations ASC

Is there a way to do this? How can I tell the query builder to exclude the "id" (primary key) attribute from the Reservations table?
I know I can solve this using raw query.

Start by reproducing the shown Reservation.findAll query with its attributes, include, group, and order options, then inspect the generated SQL. Done means the query builder can omit reservation.id from SELECT for this grouped aggregate query and the resulting SQL executes without the GROUP BY error.