diff --git a/antarest/study/business/arrow_utils.py b/antarest/study/business/arrow_utils.py
new file mode 100644
index 0000000000..8f23a7b232
--- /dev/null
+++ b/antarest/study/business/arrow_utils.py
@@ -0,0 +1,47 @@
+# Copyright (c) 2024, RTE (https://www.rte-france.com)
+#
+# See AUTHORS.txt
+#
+# This Source Code Form is subject to the terms of the Mozilla Public
+# License, v. 2.0. If a copy of the MPL was not distributed with this
+# file, You can obtain one at http://mozilla.org/MPL/2.0/.
+#
+# SPDX-License-Identifier: MPL-2.0
+#
+# This file is part of the Antares project.
+import os
+import tempfile
+import typing as t
+from io import BytesIO
+
+import pandas as pd
+import pyarrow as pa
+from pyarrow import feather
+from pyarrow.feather import write_feather
+
+
+def dataframe_to_bytes(df: pd.DataFrame, metadata: t.Optional[t.Dict[str | bytes, str | bytes]]) -> bytes:
+    table: pa.Table = pa.Table.from_pandas(df, preserve_index=False)
+
+    if metadata:
+        metadata_bytes = {str(k): str(v) for k, v in metadata.items()}
+        schema_metadata: t.Dict[str | bytes, str | bytes] = {k: v for k, v in metadata_bytes.items()}
+        table = table.replace_schema_metadata(schema_metadata)
+
+    buffer = BytesIO()
+    write_feather(df=table, dest=buffer)  # type:ignore
+
+    return buffer.getvalue()
+
+
+def bytes_to_dataframe(buffer: bytes) -> pd.DataFrame:
+    data = BytesIO(buffer)
+    table = feather.read_table(data)
+
+    df = table.to_pandas()
+
+    metadata = table.schema.metadata
+    if metadata:
+        df.metadata = {k.decode("utf8"): v.decode("utf8") for k, v in metadata.items()}
+
+    return df
diff --git a/antarest/study/business/load_management.py b/antarest/study/business/load_management.py
new file mode 100644
index 0000000000..fce3d33a23
--- /dev/null
+++ b/antarest/study/business/load_management.py
@@ -0,0 +1,55 @@
+# Copyright (c) 2024, RTE (https://www.rte-france.com)
+#
+# See AUTHORS.txt
+#
+# This Source Code Form is subject to the terms of the Mozilla Public
+# License, v. 2.0. If a copy of the MPL was not distributed with this
+# file, You can obtain one at http://mozilla.org/MPL/2.0/.
+#
+# SPDX-License-Identifier: MPL-2.0
+#
+# This file is part of the Antares project.
+
+import typing as t
+
+import pandas as pd
+
+from antarest.study.model import MatrixIndex, Study
+from antarest.study.storage.rawstudy.model.filesystem.matrix.input_series_matrix import InputSeriesMatrix
+from antarest.study.storage.storage_service import StudyStorageService
+from antarest.study.storage.utils import get_start_date
+
+LOAD_PATH = "input/load/series/load_{area_id}"
+matrix_columns = ["ts-0"]
+
+
+class LoadManager:
+    def __init__(self, storage_service: StudyStorageService) -> None:
+        self.storage_service = storage_service
+
+    def get_load_matrix(self, study: Study, area_id: str) -> t.Tuple[pd.DataFrame, t.Dict[str | bytes, str | bytes]]:
+        file_study = self.storage_service.get_storage(study).get_raw(study)
+        load_path = LOAD_PATH.format(area_id=area_id).split("/")
+
+        node = file_study.tree.get_node(load_path)
+
+        if not isinstance(node, InputSeriesMatrix):
+            raise TypeError(f"Expected node of type 'InputSeriesMatrix', but got '{type(node).__name__}'")
+
+        matrix_data = InputSeriesMatrix.parse(node, return_dataframe=True)
+
+        matrix_df = t.cast(pd.DataFrame, matrix_data)
+        matrix_df.columns = matrix_df.columns.map(str)
+
+        matrix_df.columns = pd.Index(matrix_columns)
+
+        matrix_index: MatrixIndex = get_start_date(file_study)
+
+        metadata: t.Dict[str | bytes, str | bytes] = {
+            "start_date": str(matrix_index.start_date),
+            "steps": str(matrix_index.steps),
+            "first_week_size": str(matrix_index.first_week_size),
+            "level": str(matrix_index.level),
+        }
+
+        return matrix_df, metadata
diff --git a/antarest/study/model.py b/antarest/study/model.py
index b8378aa356..15fffc8f1e 100644
--- a/antarest/study/model.py
+++ b/antarest/study/model.py
@@ -16,6 +16,7 @@
 import typing as t
 import uuid
 from datetime import datetime, timedelta
+from enum import StrEnum
 from pathlib import Path
 
 from antares.study.version import StudyVersion
@@ -521,6 +522,11 @@ def suffix(self) -> str:
         return mapping[self]
 
 
+class MatrixFormat(StrEnum):
+    JSON = "json"
+    ARROW = "arrow"
+
+
 class StudyDownloadDTO(AntaresBaseModel):
     """
     DTO used to download outputs
diff --git a/antarest/study/service.py b/antarest/study/service.py
index 7e1198c6b7..3b148daec9 100644
--- a/antarest/study/service.py
+++ b/antarest/study/service.py
@@ -89,6 +89,7 @@
 from antarest.study.business.district_manager import DistrictManager
 from antarest.study.business.general_management import GeneralManager
 from antarest.study.business.link_management import LinkManager
+from antarest.study.business.load_management import LoadManager
 from antarest.study.business.matrix_management import MatrixManager, MatrixManagerError
 from antarest.study.business.model.link_model import LinkBaseDTO, LinkDTO
 from antarest.study.business.optimization_management import OptimizationManager
@@ -394,6 +395,7 @@ def __init__(
         self.adequacy_patch_manager = AdequacyPatchManager(self.storage_service)
         self.advanced_parameters_manager = AdvancedParamsManager(self.storage_service)
         self.hydro_manager = HydroManager(self.storage_service)
+        self.load_manager = LoadManager(self.storage_service)
         self.allocation_manager = AllocationManager(self.storage_service)
         self.properties_manager = PropertiesManager(self.storage_service)
         self.renewable_manager = RenewableManager(self.storage_service)
diff --git a/antarest/study/web/study_data_blueprint.py b/antarest/study/web/study_data_blueprint.py
index 9f267df774..a166e1de56 100644
--- a/antarest/study/web/study_data_blueprint.py
+++ b/antarest/study/web/study_data_blueprint.py
@@ -17,7 +17,7 @@
 
 import typing_extensions as te
 from fastapi import APIRouter, Body, Depends, Query
-from starlette.responses import RedirectResponse
+from starlette.responses import RedirectResponse, Response
 
 from antarest.core.config import Config
 from antarest.core.jwt import JWTUser
@@ -53,6 +53,7 @@
     ThermalClusterOutput,
     ThermalManager,
 )
+from antarest.study.business.arrow_utils import dataframe_to_bytes
 from antarest.study.business.binding_constraint_management import (
     ConstraintCreation,
     ConstraintFilters,
@@ -543,6 +544,32 @@ def update_inflow_structure(
         study = study_service.check_study_access(uuid, StudyPermissionType.WRITE, params)
         study_service.hydro_manager.update_inflow_structure(study, area_id, values)
 
+    @bp.get(
+        "/studies/{uuid}/{area_id}/load/series",
+        tags=[APITag.study_data],
+        summary="Get load series data",
+    )
+    def get_load_series(
+        uuid: str,
+        area_id: str,
+        current_user: JWTUser = Depends(auth.get_current_user),
+    ) -> Response:
+        """Return the load matrix in ARROW format."""
+        logger.info(
+            msg=f"Getting load series data for area {area_id} of study {uuid}",
+            extra={"user": current_user.id},
+        )
+        params = RequestParameters(user=current_user)
+        study = study_service.check_study_access(uuid, StudyPermissionType.READ, params)
+
+        try:
+            df, metadata = study_service.load_manager.get_load_matrix(study, area_id)
+        except TypeError as e:
+            return Response(content=str(e), status_code=400)
+
+        buffer = dataframe_to_bytes(df, metadata)
+        return Response(content=buffer, media_type="application/vnd.apache.arrow.file")
+
     @bp.put(
         "/studies/{uuid}/matrix",
         tags=[APITag.study_data],
diff --git a/requirements.txt b/requirements.txt
index 60c782d945..5246430395 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -38,6 +38,8 @@ pandas~=2.2.3
 paramiko~=3.4.1
 plyer~=2.0.0
 psycopg2-binary~=2.9.9
+pyarrow~=18.1.0
+pyarrow-stubs~=10.0.1.7
 py7zr~=0.20.6
 python-json-logger~=2.0.7
 PyYAML~=5.3.1
diff --git a/tests/integration/study_data_blueprint/test_load.py b/tests/integration/study_data_blueprint/test_load.py
new file mode 100644
index 0000000000..67e5f88fb2
--- /dev/null
+++ b/tests/integration/study_data_blueprint/test_load.py
@@ -0,0 +1,50 @@
+# Copyright (c) 2024, RTE (https://www.rte-france.com)
+#
+# See AUTHORS.txt
+#
+# This Source Code Form is subject to the terms of the Mozilla Public
+# License, v. 2.0. If a copy of the MPL was not distributed with this
+# file, You can obtain one at http://mozilla.org/MPL/2.0/.
+#
+# SPDX-License-Identifier: MPL-2.0
+#
+# This file is part of the Antares project.
+from io import BytesIO
+
+import pytest
+from starlette.testclient import TestClient
+
+from antarest.study.business.arrow_utils import bytes_to_dataframe
+from tests.integration.prepare_proxy import PreparerProxy
+
+
+@pytest.mark.unit_test
+class TestLoad:
+    @pytest.mark.parametrize("study_type", ["raw", "variant"])
+    def test_load(self, client: TestClient, user_access_token: str, study_type: str) -> None:
+        client.headers = {"Authorization": f"Bearer {user_access_token}"}  # type: ignore
+
+        preparer = PreparerProxy(client, user_access_token)
+        study_id = preparer.create_study("foo", version=880)
+
+        if study_type == "variant":
+            study_id = preparer.create_variant(study_id, name="Variant 1")
+
+        area1_id = preparer.create_area(study_id, name="Area1")["id"]
+
+        # Test simple get ARROW
+
+        res = client.get(f"/v1/studies/{study_id}/{area1_id}/load/series")
+        assert res.status_code == 200
+        assert res.headers["content-type"] == "application/vnd.apache.arrow.file"
+
+        df = bytes_to_dataframe(res.content)
+
+        column_name = ["ts-0"]
+        assert column_name == list(df.columns)
+        assert df.metadata == {
+            "first_week_size": "7",
+            "level": "hourly",
+            "start_date": "2018-01-01 00:00:00",
+            "steps": "8760",
+        }