diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
index 09b9d47..2f3bac8 100644
--- a/.pre-commit-config.yaml
+++ b/.pre-commit-config.yaml
@@ -3,7 +3,7 @@
 
 repos:
 -   repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v5.0.0
+    rev: v6.0.0
     hooks:
     -   id: check-ast
     -   id: check-toml
@@ -18,7 +18,7 @@ repos:
 #    -   id: no-commit-to-branch
 
 -   repo: https://github.com/pre-commit/mirrors-mypy
-    rev: 'v1.11.2'
+    rev: 'v1.19.1'
     hooks:
     -   id: mypy
         additional_dependencies:
@@ -26,7 +26,7 @@ repos:
             - pytest
 
 -   repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: 'v0.6.9'
+    rev: 'v0.14.10'
     hooks:
     -   id: ruff
         args: [--fix]
diff --git a/pyproject.toml b/pyproject.toml
index a7bff61..25f459b 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -54,8 +54,6 @@ line-length = 100
 # The D (doc) and DTZ (datetime zone) lint classes current heavily violated - fix later
 select = ["ALL"]
 ignore = [
-    "ANN101",  # style choice - no annotation for self
-    "ANN102",  # style choice - no annotation for cls
     "CPY",  # we do not require copyright in every file
     "D",  # todo: docstring linting
     "D203",
@@ -63,8 +61,8 @@ ignore = [
     "D213",
     "DTZ", # To add
     # Linter does not detect when types are used for Pydantic
-    "TCH001",
-    "TCH003",
+    "TC001",
+    "TC003",
 ]
 
 [tool.ruff.lint.per-file-ignores]
diff --git a/src/config.py b/src/config.py
index 8a19f04..0712271 100644
--- a/src/config.py
+++ b/src/config.py
@@ -22,11 +22,11 @@ def _apply_defaults_to_siblings(configuration: TomlTable) -> TomlTable:
 
 @functools.cache
 def _load_configuration(file: Path) -> TomlTable:
-    return typing.cast(TomlTable, tomllib.loads(file.read_text()))
+    return tomllib.loads(file.read_text())
 
 
 def load_routing_configuration(file: Path = CONFIG_PATH) -> TomlTable:
-    return typing.cast(TomlTable, _load_configuration(file)["routing"])
+    return typing.cast("TomlTable", _load_configuration(file)["routing"])
 
 
 @functools.cache
diff --git a/src/database/evaluations.py b/src/database/evaluations.py
index f98b15e..799d411 100644
--- a/src/database/evaluations.py
+++ b/src/database/evaluations.py
@@ -9,7 +9,7 @@
 
 def get_math_functions(function_type: str, connection: Connection) -> Sequence[Row]:
     return cast(
-        Sequence[Row],
+        "Sequence[Row]",
         connection.execute(
             text(
                 """
diff --git a/src/database/flows.py b/src/database/flows.py
index 93fb219..3129e91 100644
--- a/src/database/flows.py
+++ b/src/database/flows.py
@@ -6,7 +6,7 @@
 
 def get_subflows(for_flow: int, expdb: Connection) -> Sequence[Row]:
     return cast(
-        Sequence[Row],
+        "Sequence[Row]",
         expdb.execute(
             text(
                 """
@@ -36,7 +36,7 @@ def get_tags(flow_id: int, expdb: Connection) -> list[str]:
 
 def get_parameters(flow_id: int, expdb: Connection) -> Sequence[Row]:
     return cast(
-        Sequence[Row],
+        "Sequence[Row]",
         expdb.execute(
             text(
                 """
diff --git a/src/database/studies.py b/src/database/studies.py
index 848c034..35c1b79 100644
--- a/src/database/studies.py
+++ b/src/database/studies.py
@@ -43,7 +43,7 @@ def get_study_data(study: Row, expdb: Connection) -> Sequence[Row]:
     """
     if study.type_ == StudyType.TASK:
         return cast(
-            Sequence[Row],
+            "Sequence[Row]",
             expdb.execute(
                 text(
                     """
@@ -56,7 +56,7 @@ def get_study_data(study: Row, expdb: Connection) -> Sequence[Row]:
             ).all(),
         )
     return cast(
-        Sequence[Row],
+        "Sequence[Row]",
         expdb.execute(
             text(
                 """
@@ -103,7 +103,7 @@ def create(study: CreateStudy, user: User, expdb: Connection) -> int:
         },
     )
     (study_id,) = expdb.execute(text("""SELECT LAST_INSERT_ID();""")).one()
-    return cast(int, study_id)
+    return cast("int", study_id)
 
 
 def attach_task(task_id: int, study_id: int, user: User, expdb: Connection) -> None:
diff --git a/src/database/tasks.py b/src/database/tasks.py
index 56a6718..97caef3 100644
--- a/src/database/tasks.py
+++ b/src/database/tasks.py
@@ -19,7 +19,7 @@ def get(id_: int, expdb: Connection) -> Row | None:
 
 def get_task_types(expdb: Connection) -> Sequence[Row]:
     return cast(
-        Sequence[Row],
+        "Sequence[Row]",
         expdb.execute(
             text(
                 """
@@ -46,7 +46,7 @@ def get_task_type(task_type_id: int, expdb: Connection) -> Row | None:
 
 def get_input_for_task_type(task_type_id: int, expdb: Connection) -> Sequence[Row]:
     return cast(
-        Sequence[Row],
+        "Sequence[Row]",
         expdb.execute(
             text(
                 """
@@ -62,7 +62,7 @@ def get_input_for_task_type(task_type_id: int, expdb: Connection) -> Sequence[Ro
 
 def get_input_for_task(id_: int, expdb: Connection) -> Sequence[Row]:
     return cast(
-        Sequence[Row],
+        "Sequence[Row]",
         expdb.execute(
             text(
                 """
@@ -78,7 +78,7 @@ def get_input_for_task(id_: int, expdb: Connection) -> Sequence[Row]:
 
 def get_task_type_inout_with_template(task_type: int, expdb: Connection) -> Sequence[Row]:
     return cast(
-        Sequence[Row],
+        "Sequence[Row]",
         expdb.execute(
             text(
                 """
diff --git a/src/routers/openml/tasks.py b/src/routers/openml/tasks.py
index 96d0198..8397f1d 100644
--- a/src/routers/openml/tasks.py
+++ b/src/routers/openml/tasks.py
@@ -1,7 +1,7 @@
 import json
 import re
 from http import HTTPStatus
-from typing import Annotated, Any
+from typing import Annotated, cast
 
 import xmltodict
 from fastapi import APIRouter, Depends, HTTPException
@@ -15,22 +15,24 @@
 
 router = APIRouter(prefix="/tasks", tags=["tasks"])
 
+type JSON = dict[str, "JSON"] | list["JSON"] | str | int | float | bool | None
 
-def convert_template_xml_to_json(xml_template: str) -> Any:  # noqa: ANN401
+
+def convert_template_xml_to_json(xml_template: str) -> dict[str, JSON]:
     json_template = xmltodict.parse(xml_template.replace("oml:", ""))
     json_str = json.dumps(json_template)
     # To account for the differences between PHP and Python conversions:
     for py, php in [("@name", "name"), ("#text", "value"), ("@type", "type")]:
         json_str = json_str.replace(py, php)
-    return json.loads(json_str)
+    return cast("dict[str, JSON]", json.loads(json_str))
 
 
 def fill_template(
     template: str,
     task: RowMapping,
-    task_inputs: dict[str, str],
+    task_inputs: dict[str, str | int],
     connection: Connection,
-) -> Any:  # noqa: ANN401
+) -> dict[str, JSON]:
     """Fill in the XML template as used for task descriptions and return the result,
      converted to JSON.
 
@@ -79,22 +81,25 @@ def fill_template(
     }
     """
     json_template = convert_template_xml_to_json(template)
-    return _fill_json_template(
-        json_template,
-        task,
-        task_inputs,
-        fetched_data={},
-        connection=connection,
+    return cast(
+        "dict[str, JSON]",
+        _fill_json_template(
+            json_template,
+            task,
+            task_inputs,
+            fetched_data={},
+            connection=connection,
+        ),
     )
 
 
 def _fill_json_template(
-    template: dict[str, Any],
+    template: JSON,
     task: RowMapping,
-    task_inputs: dict[str, str],
-    fetched_data: dict[str, Any],
+    task_inputs: dict[str, str | int],
+    fetched_data: dict[str, str],
     connection: Connection,
-) -> dict[str, Any] | list[dict[str, Any]] | str:
+) -> JSON:
     if isinstance(template, dict):
         return {
             k: _fill_json_template(v, task, task_inputs, fetched_data, connection)
@@ -115,7 +120,7 @@ def _fill_json_template(
         if match.string == template:
             # How do we know the default value? probably ttype_io table?
             return task_inputs.get(field, [])
-        template = template.replace(match.group(), task_inputs[field])
+        template = template.replace(match.group(), str(task_inputs[field]))
     if match := re.search(r"\[LOOKUP:(.*)]", template):
         (field,) = match.groups()
         if field not in fetched_data:
@@ -176,7 +181,7 @@ def get_task(
     tags = database.tasks.get_tags(task_id, expdb)
     name = f"Task {task_id} ({task_type.name})"
     dataset_id = task_inputs.get("source_data")
-    if dataset_id and (dataset := database.datasets.get(dataset_id, expdb)):
+    if isinstance(dataset_id, int) and (dataset := database.datasets.get(dataset_id, expdb)):
         name = f"Task {task_id}: {dataset.name} ({task_type.name})"
 
     return Task(
diff --git a/src/routers/openml/tasktype.py b/src/routers/openml/tasktype.py
index dcc9b1c..5213f17 100644
--- a/src/routers/openml/tasktype.py
+++ b/src/routers/openml/tasktype.py
@@ -53,11 +53,11 @@ def get_task_type(
     task_type = _normalize_task_type(task_type_record)
     # Some names are quoted, or have typos in their comma-separation (e.g. 'A ,B')
     task_type["creator"] = [
-        creator.strip(' "') for creator in cast(str, task_type["creator"]).split(",")
+        creator.strip(' "') for creator in cast("str", task_type["creator"]).split(",")
     ]
     if contributors := task_type.pop("contributors"):
         task_type["contributor"] = [
-            creator.strip(' "') for creator in cast(str, contributors).split(",")
+            creator.strip(' "') for creator in cast("str", contributors).split(",")
         ]
     task_type["creation_date"] = task_type.pop("creationDate")
     task_type_inputs = get_input_for_task_type(task_type_id, expdb)
diff --git a/src/schemas/datasets/mldcat_ap.py b/src/schemas/datasets/mldcat_ap.py
index d7e277f..ffbe644 100644
--- a/src/schemas/datasets/mldcat_ap.py
+++ b/src/schemas/datasets/mldcat_ap.py
@@ -10,7 +10,7 @@
 
 from abc import ABC
 from enum import StrEnum
-from typing import Generic, Literal, TypeVar
+from typing import Literal
 
 from pydantic import BaseModel, Field, HttpUrl, field_serializer, model_serializer
 
@@ -41,10 +41,7 @@ class JsonLDObject(BaseModel, ABC):
     }
 
 
-T = TypeVar("T", bound=JsonLDObject)
-
-
-class JsonLDObjectReference(BaseModel, Generic[T]):
+class JsonLDObjectReference[T: JsonLDObject](BaseModel):
     id_: str = Field(serialization_alias="@id")
 
     model_config = {"populate_by_name": True, "extra": "forbid"}
@@ -275,7 +272,7 @@ class DataService(JsonLDObject):
 
 
 class JsonLDGraph(BaseModel):
-    context: str | dict[str, HttpUrl] = Field(default_factory=dict, serialization_alias="@context")  # type: ignore[arg-type]
+    context: str | dict[str, HttpUrl] = Field(default_factory=dict, serialization_alias="@context")
     graph: list[Distribution | DataService | Dataset | Quality | Feature | Agent | MD5Checksum] = (
         Field(default_factory=list, serialization_alias="@graph")
     )
diff --git a/tests/routers/openml/datasets_list_datasets_test.py b/tests/routers/openml/datasets_list_datasets_test.py
index bf0fbc8..6e0492e 100644
--- a/tests/routers/openml/datasets_list_datasets_test.py
+++ b/tests/routers/openml/datasets_list_datasets_test.py
@@ -224,12 +224,12 @@ def test_list_data_quality(quality: str, range_: str, count: int, py_api: TestCl
 
 
 @pytest.mark.slow
-@hypothesis.settings(
+@hypothesis.settings(  # type: ignore[untyped-decorator]  # 108
     max_examples=5000,
     suppress_health_check=[hypothesis.HealthCheck.function_scoped_fixture],
     deadline=None,
-)  # type: ignore[misc]  # https://github.com/openml/server-api/issues/108
-@given(
+)
+@given(  # type: ignore[untyped-decorator]  # 108
     number_missing_values=st.sampled_from([None, "2", "2..10000"]),
     number_features=st.sampled_from([None, "5", "2..100"]),
     number_classes=st.sampled_from([None, "5", "2..100"]),
@@ -243,7 +243,7 @@ def test_list_data_quality(quality: str, range_: str, count: int, py_api: TestCl
     tag=st.sampled_from([None, "study_14", "study_not_in_db"]),
     # We don't test ADMIN user, as we fixed a bug which treated them as a regular user
     api_key=st.sampled_from([None, ApiKey.SOME_USER, ApiKey.OWNER_USER]),
-)  # type: ignore[misc]  # https://github.com/openml/server-api/issues/108
+)
 def test_list_data_identical(
     py_api: TestClient,
     php_api: httpx.Client,
diff --git a/tests/users.py b/tests/users.py
index ab92593..23bc325 100644
--- a/tests/users.py
+++ b/tests/users.py
@@ -9,7 +9,7 @@
 
 
 class ApiKey(StrEnum):
-    ADMIN: str = "AD000000000000000000000000000000"
-    SOME_USER: str = "00000000000000000000000000000000"
-    OWNER_USER: str = "DA1A0000000000000000000000000000"
-    INVALID: str = "11111111111111111111111111111111"
+    ADMIN = "AD000000000000000000000000000000"
+    SOME_USER = "00000000000000000000000000000000"
+    OWNER_USER = "DA1A0000000000000000000000000000"
+    INVALID = "11111111111111111111111111111111"