[Bug]: structure mp-13 returns weird primitive cell inconsistent with that shown on the webpage

[wavyparticle]

Code snippet

def get_structure_by_id(material_id:str, name:str=None):
    """
    Get the structure of a material by its ID.
    
    Parameters:
    - material_id: The ID of the material to query.
    
    Returns:
    - The structure of the material and its cif file.
    """
    with MPRester(mp_api_key) as mpr:
        structure = mpr.get_structure_by_material_id(material_id, conventional_unit_cell=False)
        
        structure.to(f"{name}_{material_id}.cif" if name else f"{material_id}.cif")
        
    print(f"Structure for {material_id} saved as {name}_{material_id}.cif" if name else f"Structure for {material_id} saved as {material_id}.cif")
        
get_structure_by_id("mp-13", name="Fe")

What happened?

for some reason the c axis from the queried structure (a primitive cell) is 4.05429521 and gamma is 109.46983668, which is inconsistent with those shown here (https://next-gen.materialsproject.org/materials/mp-13?chemsys=Fe)

with conventional_unit_cell=False:

with conventional_unit_cell=True:

right now my workaround is to query the conventional cell and convert it to primitive after which is fine, but figured I should also report the issue

screenshot from page of mp-13

conventional cell

primitive cell

also, if I try to download the primitive cell manually from the website it gives the conventional cell

appreciate any help with this; thanks!

Version

mp-api==0.45.13, pymatgen==2025.10.7

Which OS?

• MacOS
• Windows
• Linux

Log output

[esoteric-ephemera]

This is a question we get a lot - see this forum post for example

The website defaults to conventional cell representation and the asymmetric lattice angles are a non-standard way of representing the primitive cell