> For the complete documentation index, see [llms.txt](https://docs.materialsproject.org/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.materialsproject.org/downloading-data/using-the-api/legacy-api-data.md).

# Legacy API data

### Accessing Legacy Materials Project Structures

\
Older Materials Project (MP) IDs referenced in the literature and MPContribs entries may not be directly accessible via the new API, even with `deprecated=True`. Many of these structures can still be retrieved using the new API's *tasks* endpoint. The legacy API is also still accessible (as of mid-2025), but its use is discouraged for new work.

#### Recommended: Using the Next-Gen API (`mp-api`) — Tasks Endpoint

To retrieve structures for legacy MPIDs (including `mvc-` and old `mp-` IDs), use the `tasks` endpoint:

```python
from mp_api.client import MPRester

with MPRester() as mpr:
    tasks = mpr.materials.tasks.search(task_ids=['mvc-13350', 'mvc-16068', 'mvc-13133', 'mp-867303']) # Each item in 'tasks' contains the structure and associated information for task in tasks: print(task.task_id, task.structure)
```

**Note:**

* The structure you retrieve may not exactly match the one referenced in older publications.
* The `tasks` endpoint is preferred and supported for retrieving historical structure data.

#### Legacy API (not recommended, will cease operation in September 2025)

If you must access the legacy API (e.g., for reproducibility), you can use an older version of pymatgen:

1. Install pymatgen 2022.7.25 (or similar):

   ```bash
   pip install 'pymatgen==2022.7.25'
   ```
2. Query legacy structures (requires legacy API key):

   ```python
   from pymatgen.ext.matproj import MPRester as MPResterLegacy 
   from pymatgen.core import Structure 

   LEGACY_API_KEY = "your_legacy_api_key_here"  # 19-character string 
   query = {"formula": "FeO"}  # Or any other valid query 
   with MPResterLegacy(LEGACY_API_KEY) as mpr:
       materials = mpr.query(query, ["material_id", "cif"], mp_decode=False) 
       structures_by_mpid = { doc["material_id"] : Structure.from_str(doc["cif"], fmt="cif") for doc in materials }
   ```

**Caveats:**

* The legacy API is currently being phased out, and will be retired in September 2025
* There is no official support for the legacy API
* Bulk download of the full legacy database is available as of snapshot [28 October, 2022](https://materialsproject-build.s3.amazonaws.com/index.html#collections/2022-10-28/), and older snapshots may be made available (see [MP AWS downloads](https://materialsproject.org/downloads) for updates).

\
\\