`chunks` argument is typed incorrected in `open_zarr`

What happened?

If you try to type check a project that is using xarray there will be a failure when checking open_zarr. It is a best practice not to ignore this since if you do you generally have to ignore all type checking info on the open call which is otherwise useful.

This error I believe is since the default argument is of type Default which is a dummy type and thus causes issues when checking it. This essentially narrows the type more strict than just Unknown to my understanding.

Log message:

error: Argument of type "Literal['auto']" cannot be assigned to parameter "chunks" of type "Default" in function "open_zarr"

What did you expect to happen?

Open zarr should be typed in such a way that match the input arguments or not have any type information and be specified as the Unknown type

Minimal Complete Verifiable Example

# /// script
# requires-python = ">=3.11"
# dependencies = [
#   "xarray[complete]@git+https://github.com/pydata/xarray.git@main",
# ]
# ///
#
# This script automatically imports the development branch of xarray to check for issues.
# Please delete this header if you have _not_ tested this script with `uv run`!

import inspect
import xarray as xr

xr.show_versions()

from xarray import open_zarr

sig = inspect.signature(open_zarr)
store_argument = sig.parameters["store"]

# should show no type information; different type checkers will narrow 
# this differently but many including pyright / pylance will fail on it
print(store_argument.annotation)

Steps to reproduce

Type check any xarray project with pyright that uses open_zarr and passes in "auto" for the chunks arg.

MVCE confirmation

• Minimal example — the example is as focused as reasonably possible to demonstrate the underlying issue in xarray.
• Complete example — the example is self-contained, including all data and the text of any traceback.
• Verifiable example — the example copy & pastes into an IPython prompt or Binder notebook, returning the result.
• New issue — a search of GitHub Issues suggests this is not a duplicate.
• Recent environment — the issue occurs with the latest version of xarray and its dependencies.

Relevant log output

error: Argument of type "Literal['auto']" cannot be assigned to parameter "chunks" of type "Default" in function "open_zarr"

Anything else we need to know?

From the type info in the docstring it seems like the type should be

Literal["auto"] | None | Literal[-1] | dict

I can create a PR for this if needed.

I was using pyright

pyright 1.1.402

but I don't think this is specific to pyright

Environment

INSTALLED VERSIONS

commit: None
python: 3.12.8 (main, Dec 6 2024, 19:42:06) [Clang 18.1.8 ]
python-bits: 64
OS: Darwin
OS-release: 25.2.0
machine: arm64
processor: arm
byteorder: little
LC_ALL: None
LANG: en_US.UTF-8
LOCALE: ('en_US', 'UTF-8')
libhdf5: 1.14.6
libnetcdf: 4.9.3

xarray: 2026.2.0
pandas: 3.0.1
numpy: 2.4.3
scipy: None
netCDF4: 1.7.4
pydap: None
h5netcdf: None
h5py: None
zarr: 3.1.5
cftime: 1.6.5
nc_time_axis: None
iris: None
bottleneck: None
dask: None
distributed: None
matplotlib: None
cartopy: None
seaborn: None
numbagg: None
fsspec: 2026.2.0
cupy: None
pint: None
sparse: None
flox: None
numpy_groupies: None
setuptools: None
pip: None
conda: None
pytest: 9.0.2
mypy: None
IPython: None
sphinx: None

Start at the xarray.open_zarr entry point and inspect the chunks annotation and the Default type described in the issue. Reproduce the pyright failure with the provided minimal example, then verify that passing "auto" to chunks no longer produces the reported type error.