diff --git a/source/isaaclab_rl/changelog.d/default-newton-rsl.minor.rst b/source/isaaclab_rl/changelog.d/default-newton-rsl.minor.rst
new file mode 100644
index 000000000000..20e6f71b6477
--- /dev/null
+++ b/source/isaaclab_rl/changelog.d/default-newton-rsl.minor.rst
@@ -0,0 +1,6 @@
+Changed
+^^^^^^^
+
+* **Breaking:** Changed the ``train`` and ``play`` CLI commands to use a task's
+  registered default RL library when ``--rl_library`` is omitted. Pass
+  ``--rl_library`` explicitly to select a different library.
diff --git a/source/isaaclab_rl/isaaclab_rl/entrypoints/dispatch.py b/source/isaaclab_rl/isaaclab_rl/entrypoints/dispatch.py
index 67a10732880a..1a7a827c7ac6 100644
--- a/source/isaaclab_rl/isaaclab_rl/entrypoints/dispatch.py
+++ b/source/isaaclab_rl/isaaclab_rl/entrypoints/dispatch.py
@@ -13,6 +13,8 @@
 import sys
 from typing import TYPE_CHECKING
 
+import gymnasium as gym
+
 if TYPE_CHECKING:
     from .simple_agents import PolicyName
 
@@ -106,6 +108,8 @@ def run_cli(action: str, argv: list[str] | None = None) -> int:
     parser = argparse.ArgumentParser(add_help=False)
     parser.add_argument("--rl_library", choices=sorted(backends))
     selected, backend_argv = parser.parse_known_args(argv)
+    if selected.rl_library is None:
+        selected.rl_library = _resolve_default_library(argv, backends)
     if selected.rl_library is None:
         _print_selector_help(action, sorted(backends))
         if "-h" in argv or "--help" in argv:
@@ -116,6 +120,23 @@ def run_cli(action: str, argv: list[str] | None = None) -> int:
     return 0
 
 
+def _resolve_default_library(argv: list[str], backends: dict[str, str]) -> str | None:
+    """Return the task-registered default RL library requested by command-line arguments."""
+    parser = argparse.ArgumentParser(add_help=False)
+    parser.add_argument("--task")
+    args, _ = parser.parse_known_args(argv)
+    if args.task is None:
+        return None
+
+    import isaaclab_tasks  # noqa: F401
+
+    try:
+        default_library = gym.spec(args.task.split(":")[-1]).kwargs.get("default_agent")
+    except gym.error.Error:
+        return None
+    return default_library if default_library in backends else None
+
+
 def _print_selector_help(action: str, backends: list[str]) -> None:
     """Print help for a unified entrypoint before a backend is selected."""
     parser = argparse.ArgumentParser(description=f"{action.capitalize()} an RL agent with a selected backend.")
diff --git a/source/isaaclab_rl/test/test_entrypoints.py b/source/isaaclab_rl/test/test_entrypoints.py
index b6f94ed01341..d123c16d3713 100644
--- a/source/isaaclab_rl/test/test_entrypoints.py
+++ b/source/isaaclab_rl/test/test_entrypoints.py
@@ -12,6 +12,7 @@
 import sys
 import types
 
+import gymnasium as gym
 import pytest
 
 from isaaclab_rl.entrypoints import PlaybackRequest, TrainingRequest, api, dispatch
@@ -128,6 +129,32 @@ def _fake_run_backend(module_name: str, argv: list[str], *, run_as_script: bool)
     }
 
 
+def test_dispatch_uses_task_registered_default_backend(monkeypatch) -> None:
+    """A task registry default selects the backend when the CLI omits it."""
+    task_name = "Isaac-DefaultAgentDispatchTest"
+    gym.register(id=task_name, entry_point="dummy:Env", kwargs={"default_agent": "rsl_rl"})
+    monkeypatch.setitem(sys.modules, "isaaclab_tasks", types.ModuleType("isaaclab_tasks"))
+    received: dict[str, object] = {}
+    monkeypatch.setattr(
+        dispatch,
+        "_run_backend",
+        lambda module_name, argv, *, run_as_script: received.update(
+            module_name=module_name, argv=argv, run_as_script=run_as_script
+        ),
+    )
+
+    try:
+        assert dispatch.run_train_cli(["--task", task_name]) == 0
+    finally:
+        gym.registry.pop(task_name, None)
+
+    assert received == {
+        "module_name": "isaaclab_rl.entrypoints.backends.train_rsl_rl",
+        "argv": ["--task", task_name],
+        "run_as_script": False,
+    }
+
+
 def test_dispatch_fuses_option_like_kit_args(monkeypatch) -> None:
     """Space-separated option-like Kit arguments are fused before backend parsing."""
     received: dict[str, object] = {}
@@ -142,7 +169,7 @@ def test_dispatch_fuses_option_like_kit_args(monkeypatch) -> None:
 
 def test_dispatch_requires_a_backend() -> None:
     """Missing backend selection returns the conventional CLI error status."""
-    assert dispatch.run_train_cli(["--task", "Isaac-Cartpole"]) == 2
+    assert dispatch.run_train_cli([]) == 2
 
 
 def _torch_backend_state() -> tuple[bool, bool, bool, bool]:
diff --git a/source/isaaclab_tasks/changelog.d/default-newton-rsl.minor.rst b/source/isaaclab_tasks/changelog.d/default-newton-rsl.minor.rst
new file mode 100644
index 000000000000..eddf5e08deeb
--- /dev/null
+++ b/source/isaaclab_tasks/changelog.d/default-newton-rsl.minor.rst
@@ -0,0 +1,10 @@
+Changed
+^^^^^^^
+
+* **Breaking:** Changed the default task physics and renderer presets to Newton
+  MJWarp and the Newton renderer, and changed the default RL library to RSL-RL.
+  Select an explicit preset or pass ``--rl_library`` to retain a different
+  backend or RL library.
+* Changed the recorded robot-PoV camera used by the XR reference tasks to pin the
+  Isaac RTX renderer, since the Newton renderer cannot load the UDIM textures of
+  those robot assets. Select ``newton_renderer`` explicitly to override.
diff --git a/source/isaaclab_tasks/isaaclab_tasks/contrib/robot_pov_camera_cfg.py b/source/isaaclab_tasks/isaaclab_tasks/contrib/robot_pov_camera_cfg.py
index 9e57a099f5e0..dd8dc62b1fb2 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/contrib/robot_pov_camera_cfg.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/contrib/robot_pov_camera_cfg.py
@@ -5,6 +5,8 @@
 
 """Shared recorded robot-PoV camera configuration for contributed tasks."""
 
+from isaaclab_physx.renderers import IsaacRtxRendererCfg
+
 import isaaclab.sim as sim_utils
 from isaaclab.sensors import CameraCfg
 
@@ -19,6 +21,10 @@ def robot_pov_camera_cfg(
 ) -> CameraCfg:
     """Return a recorded robot-PoV camera under a prim that follows physical-body motion.
 
+    The camera defaults to the Isaac RTX renderer instead of the ``MultiBackendRendererCfg``
+    default: the Newton warp renderer cannot load the UDIM textures used by the humanoid
+    assets these feeds record. Select ``newton_renderer`` explicitly to override.
+
     Args:
         parent_prim_path: Path of the robot prim whose transform inherits physical-body motion.
         offset_pos: Camera position in the parent body frame.
@@ -30,7 +36,7 @@ def robot_pov_camera_cfg(
         height=450,
         width=720,
         data_types=["rgb"],
-        renderer_cfg=MultiBackendRendererCfg(),
+        renderer_cfg=MultiBackendRendererCfg(default=IsaacRtxRendererCfg()),
         spawn=sim_utils.PinholeCameraCfg(focal_length=18.15, clipping_range=(0.1, 2.0)),
         offset=CameraCfg.OffsetCfg(pos=offset_pos, rot=offset_rot, convention="ros"),
     )
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/cabinet/cabinet_env_cfg.py b/source/isaaclab_tasks/isaaclab_tasks/core/cabinet/cabinet_env_cfg.py
index b58fcd4aec26..d09aeec2de50 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/cabinet/cabinet_env_cfg.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/cabinet/cabinet_env_cfg.py
@@ -102,7 +102,6 @@ class CabinetSimCfg(PresetCfg):
     physx: SimulationCfg = isaacsim_physx.replace(
         physics=PhysxAutoCfg(isaacsim_physx=isaacsim_physx.physics, ovphysx=ovphysx.physics)
     )
-    default: SimulationCfg = isaacsim_physx
     newton_mjwarp: SimulationCfg = SimulationCfg(
         dt=1 / 600,
         render_interval=1,
@@ -125,6 +124,7 @@ class CabinetSimCfg(PresetCfg):
         default_visualizer_cfg=VisualizerCfg(eye=(-2.0, 2.0, 2.0), lookat=(0.8, 0.0, 0.5)),
         physics=NewtonCfg(solver_cfg=KaminoPADMMSolverCfg(max_contacts_per_world=64)),
     )
+    default: SimulationCfg = newton_mjwarp
 
 
 @configclass
@@ -138,9 +138,9 @@ class CabinetDecimationCfg(PresetCfg):
     isaacsim_physx: int = 1
     ovphysx: int = isaacsim_physx
     physx: int = isaacsim_physx
-    default: int = isaacsim_physx
     newton_mjwarp: int = 10
     newton_kamino: int = 10
+    default: int = newton_mjwarp
 
 
 ##
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/cabinet/config/franka/__init__.py b/source/isaaclab_tasks/isaaclab_tasks/core/cabinet/config/franka/__init__.py
index a2ca966c84d3..860db87e940d 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/cabinet/config/franka/__init__.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/cabinet/config/franka/__init__.py
@@ -17,6 +17,7 @@
     kwargs={
         "env_cfg_entry_point": f"{__name__}.joint_pos_env_cfg:FrankaCabinetEnvCfg",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:CabinetPPORunnerCfg",
+        "default_agent": "rsl_rl",
         "rl_games_cfg_entry_point": f"{agents.__name__}:rl_games_manager_ppo_cfg.yaml",
         "skrl_cfg_entry_point": f"{agents.__name__}:skrl_manager_ppo_cfg.yaml",
     },
@@ -35,6 +36,7 @@
         "env_cfg_entry_point": f"{__name__}.cabinet_direct_env_cfg:FrankaCabinetDirectEnvCfg",
         "rl_games_cfg_entry_point": f"{agents.__name__}:rl_games_direct_ppo_cfg.yaml",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:FrankaCabinetPPORunnerCfg",
+        "default_agent": "rsl_rl",
         "skrl_cfg_entry_point": f"{agents.__name__}:skrl_direct_ppo_cfg.yaml",
     },
 )
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/cartpole/__init__.py b/source/isaaclab_tasks/isaaclab_tasks/core/cartpole/__init__.py
index 44fe9fbfb187..064f647ecf83 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/cartpole/__init__.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/cartpole/__init__.py
@@ -37,6 +37,7 @@
         "env_cfg_entry_point": f"{__name__}.cartpole_direct_env_cfg:CartpoleEnvCfg",
         "rl_games_cfg_entry_point": f"{agents.__name__}:rl_games_direct_ppo_cfg.yaml",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:CartpoleDirectPPORunnerCfg",
+        "default_agent": "rsl_rl",
         "skrl_cfg_entry_point": f"{agents.__name__}:skrl_direct_ppo_cfg.yaml",
         "sb3_cfg_entry_point": f"{agents.__name__}:sb3_ppo_cfg.yaml",
     },
@@ -50,6 +51,7 @@
         "env_cfg_entry_point": f"{__name__}.cartpole_direct_camera_env_cfg:CartpoleCameraEnvCfg",
         "rl_games_cfg_entry_point": f"{agents.__name__}:rl_games_camera_ppo_cfg.yaml",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:CartpoleCameraDirectPPORunnerCfg",
+        "default_agent": "rsl_rl",
         "skrl_cfg_entry_point": f"{agents.__name__}:skrl_direct_camera_ppo_cfg.yaml",
     },
 )
@@ -66,6 +68,7 @@
         "env_cfg_entry_point": f"{__name__}.cartpole_manager_env_cfg:CartpoleEnvCfg",
         "rl_games_cfg_entry_point": f"{agents.__name__}:rl_games_manager_ppo_cfg.yaml",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:CartpolePPORunnerCfg",
+        "default_agent": "rsl_rl",
         "rsl_rl_with_symmetry_cfg_entry_point": (
             f"{agents.__name__}.rsl_rl_ppo_cfg:CartpolePPORunnerWithSymmetryCfg"
         ),
@@ -83,6 +86,7 @@
         "rl_games_cfg_entry_point": f"{agents.__name__}:rl_games_camera_ppo_cfg.yaml",
         "rl_games_feature_cfg_entry_point": f"{agents.__name__}:rl_games_manager_feature_ppo_cfg.yaml",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:CartpoleCameraPPORunnerCfg",
+        "default_agent": "rsl_rl",
         "rsl_rl_feature_cfg_entry_point": (
             f"{agents.__name__}.rsl_rl_ppo_cfg:CartpoleCameraFeaturePPORunnerCfg"
         ),
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/cartpole/cartpole_direct_env_cfg.py b/source/isaaclab_tasks/isaaclab_tasks/core/cartpole/cartpole_direct_env_cfg.py
index 751fc6d23ac3..9887b20e1e98 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/cartpole/cartpole_direct_env_cfg.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/cartpole/cartpole_direct_env_cfg.py
@@ -33,7 +33,6 @@ class CartpolePhysicsCfg(PresetCfg):
     isaacsim_physx: PhysxCfg = PhysxCfg()
     ovphysx: OvPhysxCfg = OvPhysxCfg()
     physx: PhysxAutoCfg = PhysxAutoCfg(isaacsim_physx=isaacsim_physx, ovphysx=ovphysx)
-    default = isaacsim_physx
     newton_mjwarp: NewtonCfg = NewtonCfg(
         solver_cfg=MJWarpSolverCfg(
             njmax=5,
@@ -51,6 +50,7 @@ class CartpolePhysicsCfg(PresetCfg):
         debug_mode=False,
         use_cuda_graph=True,
     )
+    default = newton_mjwarp
 
 
 @configclass
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/cartpole/cartpole_manager_env_cfg.py b/source/isaaclab_tasks/isaaclab_tasks/core/cartpole/cartpole_manager_env_cfg.py
index c9570664c09d..17ea5e0f5ac6 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/cartpole/cartpole_manager_env_cfg.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/cartpole/cartpole_manager_env_cfg.py
@@ -43,7 +43,6 @@ class CartpolePhysicsCfg(PresetCfg):
     isaacsim_physx: PhysxCfg = PhysxCfg()
     ovphysx: OvPhysxCfg = OvPhysxCfg()
     physx: PhysxAutoCfg = PhysxAutoCfg(isaacsim_physx=isaacsim_physx, ovphysx=ovphysx)
-    default: PhysxCfg = isaacsim_physx
     newton_mjwarp: NewtonCfg = NewtonCfg(
         solver_cfg=MJWarpSolverCfg(
             njmax=5,
@@ -56,6 +55,7 @@ class CartpolePhysicsCfg(PresetCfg):
         debug_mode=False,
         use_cuda_graph=True,
     )
+    default: NewtonCfg = newton_mjwarp
     newton_kamino: NewtonCfg = NewtonCfg(
         solver_cfg=KaminoPADMMSolverCfg(sparse_jacobian=True),
         debug_mode=False,
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/fourbar_pole/__init__.py b/source/isaaclab_tasks/isaaclab_tasks/core/fourbar_pole/__init__.py
index b30913f7e058..82fec137b8a4 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/fourbar_pole/__init__.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/fourbar_pole/__init__.py
@@ -20,5 +20,6 @@
     kwargs={
         "env_cfg_entry_point": f"{__name__}.fourbar_pole_manager_env_cfg:FourbarPoleSwingupEnvCfg",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_manager_ppo_cfg:FourbarPolePPORunnerCfg",
+        "default_agent": "rsl_rl",
     },
 )
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/handover/__init__.py b/source/isaaclab_tasks/isaaclab_tasks/core/handover/__init__.py
index 410f83ced7a3..a4f401545fee 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/handover/__init__.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/handover/__init__.py
@@ -22,6 +22,7 @@
     kwargs={
         "env_cfg_entry_point": f"{__name__}.handover_env_cfg:HandoverEnvCfg",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:HandoverPPORunnerCfg",
+        "default_agent": "rsl_rl",
         "rl_games_cfg_entry_point": f"{agents.__name__}:rl_games_ppo_cfg.yaml",
         "skrl_cfg_entry_point": f"{agents.__name__}:skrl_ppo_cfg.yaml",
         "skrl_ippo_cfg_entry_point": f"{agents.__name__}:skrl_ippo_cfg.yaml",
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/lift/config/franka/__init__.py b/source/isaaclab_tasks/isaaclab_tasks/core/lift/config/franka/__init__.py
index a70b00cb6f65..6b677df0f69b 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/lift/config/franka/__init__.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/lift/config/franka/__init__.py
@@ -20,6 +20,7 @@
     kwargs={
         "env_cfg_entry_point": f"{__name__}.franka_env_cfg:FrankaReorientEnvCfg",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:FrankaPPORunnerCfg",
+        "default_agent": "rsl_rl",
     },
 )
 
@@ -30,5 +31,6 @@
     kwargs={
         "env_cfg_entry_point": f"{__name__}.franka_env_cfg:FrankaLiftEnvCfg",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:FrankaPPORunnerCfg",
+        "default_agent": "rsl_rl",
     },
 )
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/lift/config/franka_soft/__init__.py b/source/isaaclab_tasks/isaaclab_tasks/core/lift/config/franka_soft/__init__.py
index 244d4d45718e..8f05fec805bc 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/lift/config/franka_soft/__init__.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/lift/config/franka_soft/__init__.py
@@ -18,6 +18,7 @@
     kwargs={
         "env_cfg_entry_point": f"{__name__}.franka_soft_env_cfg:FrankaSoftEnvCfg",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:FrankaDeformablePPORunnerCfg",
+        "default_agent": "rsl_rl",
     },
 )
 
@@ -28,6 +29,7 @@
     kwargs={
         "env_cfg_entry_point": f"{__name__}.franka_cloth_env_cfg:FrankaClothEnvCfg",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:FrankaClothPPORunnerCfg",
+        "default_agent": "rsl_rl",
     },
 )
 
@@ -38,6 +40,7 @@
     kwargs={
         "env_cfg_entry_point": f"{__name__}.franka_soft_env_cfg:FrankaSoftCameraEnvCfg",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:FrankaDeformableCameraPPORunnerCfg",
+        "default_agent": "rsl_rl",
     },
 )
 
@@ -48,5 +51,6 @@
     kwargs={
         "env_cfg_entry_point": f"{__name__}.franka_cloth_env_cfg:FrankaClothCameraEnvCfg",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:FrankaDeformableCameraPPORunnerCfg",
+        "default_agent": "rsl_rl",
     },
 )
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/lift/config/franka_soft/franka_cloth_env_cfg.py b/source/isaaclab_tasks/isaaclab_tasks/core/lift/config/franka_soft/franka_cloth_env_cfg.py
index 52efa6ce6cd3..77e77eae90d1 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/lift/config/franka_soft/franka_cloth_env_cfg.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/lift/config/franka_soft/franka_cloth_env_cfg.py
@@ -10,6 +10,7 @@
 from isaaclab_newton.physics import MJWarpSolverCfg, NewtonCfg, NewtonCollisionPipelineCfg
 from isaaclab_newton.sim.schemas import NewtonDeformableBodyPropertiesCfg
 from isaaclab_newton.sim.spawners.materials import NewtonSurfaceDeformableBodyMaterialCfg
+from isaaclab_physx.renderers import IsaacRtxRendererCfg
 
 import isaaclab.sim as sim_utils
 from isaaclab.assets import AssetBaseCfg, RigidObjectCfg
@@ -24,6 +25,7 @@
 from isaaclab_contrib.deformable.newton_manager_cfg import NewtonModelCfg, VBDSolverCfg
 
 from isaaclab_tasks.utils import PresetCfg
+from isaaclab_tasks.utils.presets import MultiBackendRendererCfg
 
 from ... import mdp
 from .franka_soft_env_cfg import (
@@ -171,7 +173,9 @@ class FrankaClothScenePresetCfg(PresetCfg):
 class FrankaClothCameraSceneCfg(FrankaClothSceneCfg):
     """Franka cloth scene with a base camera."""
 
-    base_camera: CameraCfg = FRANKA_CAMERA_CFG
+    base_camera: CameraCfg = FRANKA_CAMERA_CFG.replace(
+        renderer_cfg=MultiBackendRendererCfg(default=IsaacRtxRendererCfg())
+    )
 
 
 @configclass
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/lift/config/kuka_allegro/__init__.py b/source/isaaclab_tasks/isaaclab_tasks/core/lift/config/kuka_allegro/__init__.py
index 797027f05667..7f68ea22e3e5 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/lift/config/kuka_allegro/__init__.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/lift/config/kuka_allegro/__init__.py
@@ -20,6 +20,7 @@
     kwargs={
         "env_cfg_entry_point": f"{__name__}.kuka_allegro_env_cfg:KukaAllegroReorientEnvCfg",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:KukaAllegroPPORunnerCfg",
+        "default_agent": "rsl_rl",
     },
 )
 
@@ -30,6 +31,7 @@
     kwargs={
         "env_cfg_entry_point": f"{__name__}.kuka_allegro_env_cfg:KukaAllegroLiftEnvCfg",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:KukaAllegroPPORunnerCfg",
+        "default_agent": "rsl_rl",
     },
 )
 
@@ -44,6 +46,7 @@
     kwargs={
         "env_cfg_entry_point": f"{__name__}.kuka_allegro_camera_env_cfg:KukaAllegroReorientCameraEnvCfg",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:KukaAllegroPPORunnerCfg",
+        "default_agent": "rsl_rl",
     },
 )
 
@@ -54,5 +57,6 @@
     kwargs={
         "env_cfg_entry_point": f"{__name__}.kuka_allegro_camera_env_cfg:KukaAllegroLiftCameraEnvCfg",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:KukaAllegroPPORunnerCfg",
+        "default_agent": "rsl_rl",
     },
 )
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/lift/config/kuka_allegro/kuka_allegro_env_cfg.py b/source/isaaclab_tasks/isaaclab_tasks/core/lift/config/kuka_allegro/kuka_allegro_env_cfg.py
index 3b751997e266..05d71e9839d6 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/lift/config/kuka_allegro/kuka_allegro_env_cfg.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/lift/config/kuka_allegro/kuka_allegro_env_cfg.py
@@ -43,7 +43,8 @@ class KukaAllegroPhysicsCfg(lift.PhysicsCfg):
         gpu_found_lost_pairs_capacity=2**26,
     )
     physx = PhysxAutoCfg(isaacsim_physx=isaacsim_physx, ovphysx=ovphysx)
-    default = isaacsim_physx
+    newton_mjwarp = lift.PhysicsCfg().newton_mjwarp
+    default = newton_mjwarp
 
 
 @configclass
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/lift/lift_env_cfg.py b/source/isaaclab_tasks/isaaclab_tasks/core/lift/lift_env_cfg.py
index 46efc8bcd5c0..551d027e3d1f 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/lift/lift_env_cfg.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/lift/lift_env_cfg.py
@@ -509,7 +509,7 @@ class PhysicsCfg(PresetCfg):
         debug_mode=False,
     )
     physx = PhysxAutoCfg(isaacsim_physx=isaacsim_physx)
-    default = isaacsim_physx
+    default = newton_mjwarp
 
 
 @configclass
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/locomotion/ant/__init__.py b/source/isaaclab_tasks/isaaclab_tasks/core/locomotion/ant/__init__.py
index 4adb37ca8eb6..fe352c446ace 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/locomotion/ant/__init__.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/locomotion/ant/__init__.py
@@ -27,6 +27,7 @@
         "env_cfg_entry_point": f"{__name__}.ant_direct_env_cfg:AntEnvCfg",
         "rl_games_cfg_entry_point": f"{agents.__name__}:rl_games_direct_ppo_cfg.yaml",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:AntDirectPPORunnerCfg",
+        "default_agent": "rsl_rl",
         "skrl_cfg_entry_point": f"{agents.__name__}:skrl_direct_ppo_cfg.yaml",
     },
 )
@@ -42,6 +43,7 @@
     kwargs={
         "env_cfg_entry_point": f"{__name__}.ant_manager_env_cfg:AntEnvCfg",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:AntPPORunnerCfg",
+        "default_agent": "rsl_rl",
         "rl_games_cfg_entry_point": f"{agents.__name__}:rl_games_manager_ppo_cfg.yaml",
         "skrl_cfg_entry_point": f"{agents.__name__}:skrl_manager_ppo_cfg.yaml",
         "sb3_cfg_entry_point": f"{agents.__name__}:sb3_manager_ppo_cfg.yaml",
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/locomotion/ant/ant_direct_env_cfg.py b/source/isaaclab_tasks/isaaclab_tasks/core/locomotion/ant/ant_direct_env_cfg.py
index 9b482365e4ee..7c1dbf2805c6 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/locomotion/ant/ant_direct_env_cfg.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/locomotion/ant/ant_direct_env_cfg.py
@@ -33,7 +33,6 @@ class AntPhysicsCfg(PresetCfg):
     isaacsim_physx: PhysxCfg = PhysxCfg(bounce_threshold_velocity=0.2)
     ovphysx: OvPhysxCfg = OvPhysxCfg()
     physx: PhysxAutoCfg = PhysxAutoCfg(isaacsim_physx=isaacsim_physx, ovphysx=ovphysx)
-    default = isaacsim_physx
     newton_mjwarp: NewtonCfg = NewtonCfg(
         solver_cfg=MJWarpSolverCfg(
             njmax=45,
@@ -50,6 +49,7 @@ class AntPhysicsCfg(PresetCfg):
         debug_mode=False,
         use_cuda_graph=True,
     )
+    default = newton_mjwarp
 
 
 @configclass
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/locomotion/ant/ant_manager_env_cfg.py b/source/isaaclab_tasks/isaaclab_tasks/core/locomotion/ant/ant_manager_env_cfg.py
index bfdc93912e6a..d628496cf9a3 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/locomotion/ant/ant_manager_env_cfg.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/locomotion/ant/ant_manager_env_cfg.py
@@ -37,7 +37,6 @@ class AntPhysicsCfg(PresetCfg):
     isaacsim_physx: PhysxCfg = PhysxCfg(bounce_threshold_velocity=0.2)
     ovphysx: OvPhysxCfg = OvPhysxCfg()
     physx: PhysxAutoCfg = PhysxAutoCfg(isaacsim_physx=isaacsim_physx, ovphysx=ovphysx)
-    default: PhysxCfg = isaacsim_physx
     newton_mjwarp: NewtonCfg = NewtonCfg(
         solver_cfg=MJWarpSolverCfg(
             njmax=45,
@@ -54,6 +53,7 @@ class AntPhysicsCfg(PresetCfg):
         debug_mode=False,
         use_cuda_graph=True,
     )
+    default: NewtonCfg = newton_mjwarp
 
 
 @configclass
@@ -143,8 +143,8 @@ def __post_init__(self):
 class AntObservationsCfg(PresetCfg):
     physx: ObservationsCfg = ObservationsCfg()
     isaacsim_physx: ObservationsCfg = physx
-    default: ObservationsCfg = isaacsim_physx
     newton_mjwarp: ObservationsCfg = ObservationsCfg()
+    default: ObservationsCfg = newton_mjwarp
 
 
 @configclass
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/locomotion/humanoid/__init__.py b/source/isaaclab_tasks/isaaclab_tasks/core/locomotion/humanoid/__init__.py
index d13b7697595a..1c43bd1e3fcb 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/locomotion/humanoid/__init__.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/locomotion/humanoid/__init__.py
@@ -27,6 +27,7 @@
         "env_cfg_entry_point": f"{__name__}.humanoid_direct_env_cfg:HumanoidEnvCfg",
         "rl_games_cfg_entry_point": f"{agents.__name__}:rl_games_direct_ppo_cfg.yaml",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:HumanoidDirectPPORunnerCfg",
+        "default_agent": "rsl_rl",
         "skrl_cfg_entry_point": f"{agents.__name__}:skrl_direct_ppo_cfg.yaml",
     },
 )
@@ -42,6 +43,7 @@
     kwargs={
         "env_cfg_entry_point": f"{__name__}.humanoid_manager_env_cfg:HumanoidEnvCfg",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:HumanoidPPORunnerCfg",
+        "default_agent": "rsl_rl",
         "rl_games_cfg_entry_point": f"{agents.__name__}:rl_games_manager_ppo_cfg.yaml",
         "skrl_cfg_entry_point": f"{agents.__name__}:skrl_manager_ppo_cfg.yaml",
         "sb3_cfg_entry_point": f"{agents.__name__}:sb3_manager_ppo_cfg.yaml",
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/locomotion/humanoid/humanoid_direct_env_cfg.py b/source/isaaclab_tasks/isaaclab_tasks/core/locomotion/humanoid/humanoid_direct_env_cfg.py
index 19fe73f607fc..b23cdd566958 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/locomotion/humanoid/humanoid_direct_env_cfg.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/locomotion/humanoid/humanoid_direct_env_cfg.py
@@ -29,7 +29,6 @@ class HumanoidPhysicsCfg(PresetCfg):
     isaacsim_physx: PhysxCfg = PhysxCfg(bounce_threshold_velocity=0.2)
     ovphysx: OvPhysxCfg = OvPhysxCfg()
     physx: PhysxAutoCfg = PhysxAutoCfg(isaacsim_physx=isaacsim_physx, ovphysx=ovphysx)
-    default = isaacsim_physx
     newton_mjwarp: NewtonCfg = NewtonCfg(
         solver_cfg=MJWarpSolverCfg(
             njmax=80,
@@ -42,6 +41,7 @@ class HumanoidPhysicsCfg(PresetCfg):
         num_substeps=2,
         debug_mode=False,
     )
+    default = newton_mjwarp
 
 
 @configclass
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/locomotion/humanoid/humanoid_manager_env_cfg.py b/source/isaaclab_tasks/isaaclab_tasks/core/locomotion/humanoid/humanoid_manager_env_cfg.py
index d47422c1ad9a..11a3aa6f16f6 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/locomotion/humanoid/humanoid_manager_env_cfg.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/locomotion/humanoid/humanoid_manager_env_cfg.py
@@ -49,7 +49,6 @@ class HumanoidPhysicsCfg(PresetCfg):
     isaacsim_physx: PhysxCfg = PhysxCfg(bounce_threshold_velocity=0.2)
     ovphysx: OvPhysxCfg = OvPhysxCfg()
     physx: PhysxAutoCfg = PhysxAutoCfg(isaacsim_physx=isaacsim_physx, ovphysx=ovphysx)
-    default: PhysxCfg = isaacsim_physx
     newton_mjwarp: NewtonCfg = NewtonCfg(
         solver_cfg=MJWarpSolverCfg(
             njmax=80,
@@ -62,6 +61,7 @@ class HumanoidPhysicsCfg(PresetCfg):
         num_substeps=2,
         debug_mode=False,
     )
+    default: NewtonCfg = newton_mjwarp
 
 
 ##
@@ -147,8 +147,8 @@ def __post_init__(self):
 class HumanoidObservationsCfg(PresetCfg):
     physx: ObservationsCfg = ObservationsCfg()
     isaacsim_physx: ObservationsCfg = physx
-    default: ObservationsCfg = isaacsim_physx
     newton_mjwarp: ObservationsCfg = ObservationsCfg()
+    default: ObservationsCfg = newton_mjwarp
 
 
 @configclass
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/pendulum/__init__.py b/source/isaaclab_tasks/isaaclab_tasks/core/pendulum/__init__.py
index 4f3071b17f7c..caeabe042613 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/pendulum/__init__.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/pendulum/__init__.py
@@ -21,6 +21,8 @@
     disable_env_checker=True,
     kwargs={
         "env_cfg_entry_point": f"{__name__}.pendulum_env_cfg:PendulumEnvCfg",
+        "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:PendulumPPORunnerCfg",
+        "default_agent": "rsl_rl",
         "rl_games_cfg_entry_point": f"{agents.__name__}:rl_games_ppo_cfg.yaml",
         "skrl_cfg_entry_point": f"{agents.__name__}:skrl_ppo_cfg.yaml",
         "skrl_ippo_cfg_entry_point": f"{agents.__name__}:skrl_ippo_cfg.yaml",
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/pendulum/agents/rsl_rl_ppo_cfg.py b/source/isaaclab_tasks/isaaclab_tasks/core/pendulum/agents/rsl_rl_ppo_cfg.py
new file mode 100644
index 000000000000..ddf7f022c745
--- /dev/null
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/pendulum/agents/rsl_rl_ppo_cfg.py
@@ -0,0 +1,44 @@
+# Copyright (c) 2022-2026, The Isaac Lab Project Developers (https://github.com/isaac-sim/IsaacLab/blob/main/CONTRIBUTORS.md).
+# All rights reserved.
+#
+# SPDX-License-Identifier: BSD-3-Clause
+
+from isaaclab.utils.configclass import configclass
+
+from isaaclab_rl.rsl_rl import RslRlMLPModelCfg, RslRlOnPolicyRunnerCfg, RslRlPpoAlgorithmCfg
+
+
+@configclass
+class PendulumPPORunnerCfg(RslRlOnPolicyRunnerCfg):
+    """RSL-RL PPO configuration for the single-agent view of cart-double-pendulum balancing."""
+
+    num_steps_per_env = 32
+    max_iterations = 500
+    save_interval = 50
+    experiment_name = "pendulum"
+    obs_groups = {"actor": ["policy"], "critic": ["policy"]}
+    actor = RslRlMLPModelCfg(
+        hidden_dims=[64, 64],
+        activation="elu",
+        obs_normalization=True,
+        distribution_cfg=RslRlMLPModelCfg.GaussianDistributionCfg(init_std=1.0),
+    )
+    critic = RslRlMLPModelCfg(
+        hidden_dims=[64, 64],
+        activation="elu",
+        obs_normalization=True,
+    )
+    algorithm = RslRlPpoAlgorithmCfg(
+        value_loss_coef=1.0,
+        use_clipped_value_loss=True,
+        clip_param=0.2,
+        entropy_coef=0.001,
+        num_learning_epochs=5,
+        num_mini_batches=4,
+        learning_rate=1.0e-3,
+        schedule="adaptive",
+        gamma=0.99,
+        lam=0.95,
+        desired_kl=0.01,
+        max_grad_norm=1.0,
+    )
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/reach/config/franka/__init__.py b/source/isaaclab_tasks/isaaclab_tasks/core/reach/config/franka/__init__.py
index e697f57dd3b2..7a2415c2b864 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/reach/config/franka/__init__.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/reach/config/franka/__init__.py
@@ -23,6 +23,7 @@
         "env_cfg_entry_point": f"{__name__}.franka_reach_env_cfg:FrankaReachEnvCfg",
         "rl_games_cfg_entry_point": f"{agents.__name__}:rl_games_ppo_cfg.yaml",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:FrankaReachPPORunnerCfg",
+        "default_agent": "rsl_rl",
         "skrl_cfg_entry_point": f"{agents.__name__}:skrl_ppo_cfg.yaml",
     },
 )
@@ -38,5 +39,6 @@
     kwargs={
         "env_cfg_entry_point": f"{__name__}.franka_reach_osc_env_cfg:FrankaReachEnvCfg",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:FrankaReachPPORunnerCfg",
+        "default_agent": "rsl_rl",
     },
 )
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/reach/config/ur_10/__init__.py b/source/isaaclab_tasks/isaaclab_tasks/core/reach/config/ur_10/__init__.py
index 50cf46a54fc1..8424eb7fc250 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/reach/config/ur_10/__init__.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/reach/config/ur_10/__init__.py
@@ -19,6 +19,7 @@
         "env_cfg_entry_point": f"{__name__}.joint_pos_env_cfg:UR10ReachEnvCfg",
         "rl_games_cfg_entry_point": f"{agents.__name__}:rl_games_ppo_cfg.yaml",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:UR10ReachPPORunnerCfg",
+        "default_agent": "rsl_rl",
         "skrl_cfg_entry_point": f"{agents.__name__}:skrl_ppo_cfg.yaml",
     },
 )
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/reorient/config/allegro_hand/__init__.py b/source/isaaclab_tasks/isaaclab_tasks/core/reorient/config/allegro_hand/__init__.py
index 8571ff7b154a..91581eec335d 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/reorient/config/allegro_hand/__init__.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/reorient/config/allegro_hand/__init__.py
@@ -21,6 +21,7 @@
         "env_cfg_entry_point": f"{__name__}.allegro_hand_direct_env_cfg:AllegroHandEnvCfg",
         "rl_games_cfg_entry_point": f"{agents.__name__}:rl_games_ppo_cfg.yaml",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:AllegroHandDirectPPORunnerCfg",
+        "default_agent": "rsl_rl",
         "skrl_cfg_entry_point": f"{agents.__name__}:skrl_ppo_cfg.yaml",
     },
 )
@@ -36,6 +37,7 @@
     kwargs={
         "env_cfg_entry_point": f"{__name__}.allegro_hand_manager_env_cfg:AllegroHandManagerEnvCfg",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:AllegroHandManagerPPORunnerCfg",
+        "default_agent": "rsl_rl",
         "rl_games_cfg_entry_point": f"{agents.__name__}:rl_games_ppo_cfg.yaml",
         "skrl_cfg_entry_point": f"{agents.__name__}:skrl_ppo_cfg.yaml",
     },
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/reorient/config/shadow_hand/__init__.py b/source/isaaclab_tasks/isaaclab_tasks/core/reorient/config/shadow_hand/__init__.py
index 2c8c1e7c364b..9f3754a27c27 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/reorient/config/shadow_hand/__init__.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/reorient/config/shadow_hand/__init__.py
@@ -26,6 +26,7 @@
         "env_cfg_entry_point": f"{__name__}.shadow_hand_direct_env_cfg:ShadowHandEnvCfg",
         "rl_games_cfg_entry_point": f"{agents.__name__}:rl_games_ppo_cfg.yaml",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:ShadowHandPPORunnerCfg",
+        "default_agent": "rsl_rl",
         "skrl_cfg_entry_point": f"{agents.__name__}:skrl_ppo_cfg.yaml",
     },
 )
@@ -37,6 +38,7 @@
         "env_cfg_entry_point": f"{__name__}.shadow_hand_manager_env_cfg:ShadowHandManagerEnvPresetCfg",
         "rl_games_cfg_entry_point": f"{agents.__name__}:rl_games_ppo_cfg.yaml",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:ShadowHandManagerPPORunnerCfg",
+        "default_agent": "rsl_rl",
         "skrl_cfg_entry_point": f"{agents.__name__}:skrl_ppo_cfg.yaml",
     },
 )
@@ -52,6 +54,7 @@
     kwargs={
         "env_cfg_entry_point": f"{__name__}.shadow_hand_direct_camera_env_cfg:ShadowHandCameraEnvCfg",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:ShadowHandCameraFFPPORunnerCfg",
+        "default_agent": "rsl_rl",
         "rl_games_cfg_entry_point": f"{agents.__name__}:rl_games_ppo_camera_cfg.yaml",
     },
 )
@@ -62,6 +65,7 @@
     kwargs={
         "env_cfg_entry_point": f"{__name__}.shadow_hand_camera_manager_env_cfg:ShadowHandCameraManagerEnvCfg",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:ShadowHandCameraFFPPORunnerCfg",
+        "default_agent": "rsl_rl",
         "rl_games_cfg_entry_point": f"{agents.__name__}:rl_games_ppo_camera_cfg.yaml",
     },
 )
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/anymal_d/__init__.py b/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/anymal_d/__init__.py
index 8447486636e5..775381be0427 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/anymal_d/__init__.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/anymal_d/__init__.py
@@ -18,6 +18,7 @@
     kwargs={
         "env_cfg_entry_point": f"{__name__}.flat_env_cfg:AnymalDFlatEnvCfg",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:AnymalDFlatPPORunnerCfg",
+        "default_agent": "rsl_rl",
         "rsl_rl_recurrent_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:AnymalDFlatPPORunnerRecurrentCfg",
         "rsl_rl_distillation_cfg_entry_point": (
             f"{agents.__name__}.rsl_rl_distillation_cfg:AnymalDFlatDistillationRunnerCfg"
@@ -37,6 +38,7 @@
     kwargs={
         "env_cfg_entry_point": f"{__name__}.rough_env_cfg:AnymalDRoughEnvCfg",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:AnymalDRoughPPORunnerCfg",
+        "default_agent": "rsl_rl",
         "rsl_rl_with_symmetry_cfg_entry_point": (
             f"{agents.__name__}.rsl_rl_ppo_cfg:AnymalDRoughPPORunnerWithSymmetryCfg"
         ),
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/anymal_d/flat_env_cfg.py b/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/anymal_d/flat_env_cfg.py
index 8b6465f7b449..f2b96e28d0b5 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/anymal_d/flat_env_cfg.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/anymal_d/flat_env_cfg.py
@@ -32,7 +32,7 @@ class PhysicsCfg(PresetCfg):
         debug_mode=False,
     )
     newton_kamino = NewtonCfg(solver_cfg=KaminoPADMMSolverCfg(max_contacts_per_world=64))
-    default = isaacsim_physx
+    default = newton_mjwarp
 
 
 @configclass
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/cassie/__init__.py b/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/cassie/__init__.py
index 7ba60236964e..ea04e8ebf93b 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/cassie/__init__.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/cassie/__init__.py
@@ -18,6 +18,7 @@
     kwargs={
         "env_cfg_entry_point": f"{__name__}.flat_env_cfg:CassieFlatEnvCfg",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:CassieFlatPPORunnerCfg",
+        "default_agent": "rsl_rl",
         "skrl_cfg_entry_point": f"{agents.__name__}:skrl_flat_ppo_cfg.yaml",
     },
 )
@@ -29,6 +30,7 @@
     kwargs={
         "env_cfg_entry_point": f"{__name__}.rough_env_cfg:CassieRoughEnvCfg",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:CassieRoughPPORunnerCfg",
+        "default_agent": "rsl_rl",
         "skrl_cfg_entry_point": f"{agents.__name__}:skrl_rough_ppo_cfg.yaml",
     },
 )
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/cassie/flat_env_cfg.py b/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/cassie/flat_env_cfg.py
index 1cf5501abfe1..21ba95740cdb 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/cassie/flat_env_cfg.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/cassie/flat_env_cfg.py
@@ -33,7 +33,7 @@ class PhysicsCfg(PresetCfg):
         debug_mode=False,
     )
     newton_kamino = NewtonCfg(solver_cfg=KaminoPADMMSolverCfg(max_contacts_per_world=64))
-    default = isaacsim_physx
+    default = newton_mjwarp
 
 
 @configclass
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/g1/__init__.py b/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/g1/__init__.py
index 6fa1a65d258e..c812894c8de7 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/g1/__init__.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/g1/__init__.py
@@ -18,6 +18,7 @@
     kwargs={
         "env_cfg_entry_point": f"{__name__}.rough_env_cfg:G1RoughEnvCfg",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:G1RoughPPORunnerCfg",
+        "default_agent": "rsl_rl",
         "skrl_cfg_entry_point": f"{agents.__name__}:skrl_rough_ppo_cfg.yaml",
     },
 )
@@ -30,6 +31,7 @@
     kwargs={
         "env_cfg_entry_point": f"{__name__}.flat_env_cfg:G1FlatEnvCfg",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:G1FlatPPORunnerCfg",
+        "default_agent": "rsl_rl",
         "skrl_cfg_entry_point": f"{agents.__name__}:skrl_flat_ppo_cfg.yaml",
     },
 )
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/g1/flat_env_cfg.py b/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/g1/flat_env_cfg.py
index b40e021d6103..5a53be0c3ee3 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/g1/flat_env_cfg.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/g1/flat_env_cfg.py
@@ -34,7 +34,7 @@ class PhysicsCfg(PresetCfg):
         debug_mode=False,
     )
     newton_kamino = NewtonCfg(solver_cfg=KaminoPADMMSolverCfg(max_contacts_per_world=64))
-    default = isaacsim_physx
+    default = newton_mjwarp
 
 
 @configclass
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/go2/__init__.py b/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/go2/__init__.py
index b51cf3b73942..4e4c8c9744a0 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/go2/__init__.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/go2/__init__.py
@@ -18,6 +18,7 @@
     kwargs={
         "env_cfg_entry_point": f"{__name__}.flat_env_cfg:UnitreeGo2FlatEnvCfg",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:UnitreeGo2FlatPPORunnerCfg",
+        "default_agent": "rsl_rl",
         "skrl_cfg_entry_point": f"{agents.__name__}:skrl_flat_ppo_cfg.yaml",
     },
 )
@@ -29,6 +30,7 @@
     kwargs={
         "env_cfg_entry_point": f"{__name__}.rough_env_cfg:UnitreeGo2RoughEnvCfg",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:UnitreeGo2RoughPPORunnerCfg",
+        "default_agent": "rsl_rl",
         "skrl_cfg_entry_point": f"{agents.__name__}:skrl_rough_ppo_cfg.yaml",
     },
 )
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/go2/flat_env_cfg.py b/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/go2/flat_env_cfg.py
index b40781cba880..bcb90551d801 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/go2/flat_env_cfg.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/go2/flat_env_cfg.py
@@ -33,7 +33,7 @@ class PhysicsCfg(PresetCfg):
         debug_mode=False,
     )
     newton_kamino = NewtonCfg(solver_cfg=KaminoPADMMSolverCfg(max_contacts_per_world=64))
-    default = isaacsim_physx
+    default = newton_mjwarp
 
 
 @configclass
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/h1/__init__.py b/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/h1/__init__.py
index 1eefd014e6f8..d2fca850d2c4 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/h1/__init__.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/h1/__init__.py
@@ -18,6 +18,7 @@
     kwargs={
         "env_cfg_entry_point": f"{__name__}.rough_env_cfg:H1RoughEnvCfg",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:H1RoughPPORunnerCfg",
+        "default_agent": "rsl_rl",
         "skrl_cfg_entry_point": f"{agents.__name__}:skrl_rough_ppo_cfg.yaml",
     },
 )
@@ -30,6 +31,7 @@
     kwargs={
         "env_cfg_entry_point": f"{__name__}.flat_env_cfg:H1FlatEnvCfg",
         "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:H1FlatPPORunnerCfg",
+        "default_agent": "rsl_rl",
         "skrl_cfg_entry_point": f"{agents.__name__}:skrl_flat_ppo_cfg.yaml",
     },
 )
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/h1/flat_env_cfg.py b/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/h1/flat_env_cfg.py
index 33b128a80632..1e77fdb16f3f 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/h1/flat_env_cfg.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/velocity/config/h1/flat_env_cfg.py
@@ -33,7 +33,7 @@ class PhysicsCfg(PresetCfg):
         debug_mode=False,
     )
     newton_kamino = NewtonCfg(solver_cfg=KaminoPADMMSolverCfg(max_contacts_per_world=64))
-    default = isaacsim_physx
+    default = newton_mjwarp
 
 
 @configclass
diff --git a/source/isaaclab_tasks/isaaclab_tasks/core/velocity/velocity_env_cfg.py b/source/isaaclab_tasks/isaaclab_tasks/core/velocity/velocity_env_cfg.py
index 4133c7d3e8f9..c6e4e199704e 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/core/velocity/velocity_env_cfg.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/core/velocity/velocity_env_cfg.py
@@ -64,7 +64,7 @@ class RoughPhysicsCfg(PresetCfg):
         debug_mode=False,
         default_shape_cfg=NewtonShapeCfg(margin=0.0, ke=160000.0, kd=1100.0),
     )
-    default = isaacsim_physx
+    default = newton_mjwarp
 
 
 ##
diff --git a/source/isaaclab_tasks/isaaclab_tasks/utils/presets.py b/source/isaaclab_tasks/isaaclab_tasks/utils/presets.py
index efa77ff12ae8..2ba5760d5b7c 100644
--- a/source/isaaclab_tasks/isaaclab_tasks/utils/presets.py
+++ b/source/isaaclab_tasks/isaaclab_tasks/utils/presets.py
@@ -22,11 +22,11 @@ class _AutoRtxRendererCfg(RendererCfg):
 
 @configclass
 class MultiBackendRendererCfg(PresetCfg):
-    default: IsaacRtxRendererCfg = IsaacRtxRendererCfg()
+    default: NewtonWarpRendererCfg = NewtonWarpRendererCfg()
     rtx: _AutoRtxRendererCfg = _AutoRtxRendererCfg()
     newton_renderer: NewtonWarpRendererCfg = NewtonWarpRendererCfg()
     ovrtx: OVRTXRendererCfg = OVRTXRendererCfg()
-    isaacsim_rtx = default
+    isaacsim_rtx: IsaacRtxRendererCfg = IsaacRtxRendererCfg()
 
 
 def set_isaac_rtx_global_settings(renderer_cfg: Any, **settings: Any) -> None:
diff --git a/source/isaaclab_tasks/test/core/test_environment_determinism.py b/source/isaaclab_tasks/test/core/test_environment_determinism.py
index 74c558fcd1ae..24deddb93ec5 100644
--- a/source/isaaclab_tasks/test/core/test_environment_determinism.py
+++ b/source/isaaclab_tasks/test/core/test_environment_determinism.py
@@ -23,7 +23,13 @@
 
 import isaaclab_tasks  # noqa: F401
 from isaaclab_tasks.utils.hydra import resolve_presets
-from isaaclab_tasks.utils.parse_cfg import load_cfg_from_registry, parse_env_cfg
+from isaaclab_tasks.utils.parse_cfg import load_cfg_from_registry
+
+# Physics preset pinned for every task below. Bit-wise determinism is a property of the physics
+# backend, so the backend is selected explicitly instead of following whichever one the task
+# currently defaults to. Newton MJWarp is not yet run-to-run deterministic on CUDA for the
+# locomotion tasks, so a default flip must not silently turn this suite into a Newton test.
+_PINNED_PHYSICS_PRESET = "isaacsim_physx"
 
 
 @pytest.fixture(scope="module", autouse=True)
@@ -100,7 +106,7 @@ def _test_environment_determinism(
     device: str,
     *,
     num_steps: int = 100,
-    physics_preset_name: str | None = None,
+    physics_preset_name: str = _PINNED_PHYSICS_PRESET,
     deterministic_mode: str | None = None,
 ):
     """Check deterministic environment creation."""
@@ -138,21 +144,18 @@ def _obtain_transition_tuples(
     device: str,
     num_steps: int,
     *,
-    physics_preset_name: str | None = None,
+    physics_preset_name: str,
     deterministic_mode: str | None = None,
 ) -> tuple[dict, torch.Tensor]:
     """Run random actions and obtain transition tuples after fixed number of steps."""
     # create a new stage
     sim_utils.create_new_stage()
     try:
-        # parse configuration
-        if physics_preset_name is None:
-            env_cfg = parse_env_cfg(task_name, device=device, num_envs=num_envs)
-        else:
-            env_cfg = load_cfg_from_registry(task_name, "env_cfg_entry_point")
-            env_cfg = resolve_presets(env_cfg, selected=(physics_preset_name,))
-            env_cfg.sim.device = device
-            env_cfg.scene.num_envs = num_envs
+        # Parse configuration with the selected physics backend pinned.
+        env_cfg = load_cfg_from_registry(task_name, "env_cfg_entry_point")
+        env_cfg = resolve_presets(env_cfg, selected=(physics_preset_name,))
+        env_cfg.sim.device = device
+        env_cfg.scene.num_envs = num_envs
         if deterministic_mode is not None:
             assert isinstance(env_cfg.sim.physics, NewtonCfg)
             env_cfg.sim.physics.deterministic_mode = deterministic_mode
diff --git a/source/isaaclab_tasks/test/core/test_preset_kit_decision.py b/source/isaaclab_tasks/test/core/test_preset_kit_decision.py
index 73a119182f3c..5b14f3c7934e 100644
--- a/source/isaaclab_tasks/test/core/test_preset_kit_decision.py
+++ b/source/isaaclab_tasks/test/core/test_preset_kit_decision.py
@@ -15,6 +15,7 @@
 
 import gymnasium as gym
 import pytest
+from isaaclab_newton.physics import NewtonCfg
 from isaaclab_ov.physics import OvPhysxCfg
 from isaaclab_ov.renderers import OVRTXRendererCfg
 from isaaclab_physx.physics import PhysxCfg
@@ -87,11 +88,12 @@ def test_isaacsim_physx_is_physics_selector():
     assert "isaacsim_physx" in preset_map[PresetTarget.PHYSICS]
 
 
-def test_registered_task_physx_presets_keep_auto_selection_explicit():
-    """PhysX defaults are concrete while ``physx`` remains the automatic selector."""
+def test_core_task_physx_presets_keep_auto_selection_explicit():
+    """Core tasks retain explicit PhysX variants alongside automatic ``physx``."""
 
     for task_id, task_spec in gym.registry.items():
-        if not task_id.startswith(("Isaac-", "IsaacContrib-")) or "env_cfg_entry_point" not in task_spec.kwargs:
+        entry_point = task_spec.kwargs.get("env_cfg_entry_point", "")
+        if not task_id.startswith("Isaac-") or "isaaclab_tasks.core" not in str(entry_point):
             continue
         env_cfg = load_cfg_from_registry(task_id, "env_cfg_entry_point")
         presets = collect_presets(env_cfg)
@@ -137,24 +139,24 @@ def test_preset_mjwarp_ovrtx_does_not_need_kit():
     assert needs_kit is False
 
 
-def test_preset_rtx_with_default_physx_resolves_to_isaac_sim_backends():
-    """Automatic RTX follows the default concrete Isaac Sim PhysX backend."""
+def test_preset_rtx_with_default_newton_resolves_to_isaac_sim_backends():
+    """Explicit RTX retains the default Newton physics while requiring Isaac Sim."""
     env_cfg = _resolve_with_presets("rtx")
     config_scan = _resolve_runtime_renderer(env_cfg)
 
-    assert isinstance(env_cfg.sim.physics, PhysxCfg)
-    assert isinstance(env_cfg.tiled_camera.renderer_cfg, IsaacRtxRendererCfg)
-    assert config_scan.needs_kit is True
+    assert isinstance(env_cfg.sim.physics, NewtonCfg)
+    assert isinstance(env_cfg.tiled_camera.renderer_cfg, OVRTXRendererCfg)
+    assert config_scan.needs_kit is False
 
 
-def test_renderer_selector_rtx_with_default_physx_resolves_to_isaac_sim_backends():
-    """The RTX selector follows the default concrete Isaac Sim PhysX backend."""
+def test_renderer_selector_rtx_with_default_newton_resolves_to_isaac_sim_backends():
+    """The RTX selector retains the default Newton physics while requiring Isaac Sim."""
     env_cfg = _resolve_with_args("renderer=rtx")
     config_scan = _resolve_runtime_renderer(env_cfg)
 
-    assert isinstance(env_cfg.sim.physics, PhysxCfg)
-    assert isinstance(env_cfg.tiled_camera.renderer_cfg, IsaacRtxRendererCfg)
-    assert config_scan.needs_kit is True
+    assert isinstance(env_cfg.sim.physics, NewtonCfg)
+    assert isinstance(env_cfg.tiled_camera.renderer_cfg, OVRTXRendererCfg)
+    assert config_scan.needs_kit is False
 
 
 def test_renderer_selector_physx_rtx_resolves_to_ovphysx_without_kit():
@@ -227,20 +229,20 @@ def test_preset_mjwarp_newton_renderer_does_not_need_kit():
     assert needs_kit is False
 
 
-def test_preset_physx_with_default_kit_camera_resolves_to_physx():
-    """Automatic PhysX resolves to Isaac Sim PhysX when the default camera requires Kit."""
+def test_preset_physx_with_default_newton_camera_resolves_to_ovphysx():
+    """Automatic PhysX resolves to OvPhysX when the default camera is kitless."""
     env_cfg = _resolve_with_presets("physx")
     config_scan = scan(env_cfg)
 
-    assert isinstance(env_cfg.sim.physics, PhysxCfg)
-    assert config_scan.needs_kit is True
+    assert isinstance(env_cfg.sim.physics, OvPhysxCfg)
+    assert config_scan.needs_kit is False
 
 
-def test_preset_default_needs_kit():
-    """Default concrete Isaac Sim PhysX plus Isaac RTX requires Kit."""
+def test_preset_default_is_kitless():
+    """Default Newton MJWarp plus Newton renderer does not require Kit."""
     env_cfg = _resolve_with_presets("default")
     needs_kit = scan(env_cfg).needs_kit
-    assert needs_kit is True
+    assert needs_kit is False
 
 
 def test_preset_mjwarp_isaac_rtx_needs_kit():
diff --git a/source/isaaclab_tasks/test/core/test_rendering_registered_tasks.py b/source/isaaclab_tasks/test/core/test_rendering_registered_tasks.py
index 9e32e8b57c1b..6c8d38e5dba4 100644
--- a/source/isaaclab_tasks/test/core/test_rendering_registered_tasks.py
+++ b/source/isaaclab_tasks/test/core/test_rendering_registered_tasks.py
@@ -61,6 +61,11 @@ def _collect_camera_outputs(env: object) -> dict[str, dict[str, torch.Tensor]]:
     return outputs
 
 
+# Physics and renderer presets pinned for every combination below. Golden stages and images are
+# backend-specific, so they are compared against an explicit backend rather than whichever backend
+# the task currently defaults to.
+_PINNED_PRESETS = ("isaacsim_physx", "isaacsim_rtx")
+
 # Task IDs that expose camera/tiled_camera image observations; each is validated for non-blank
 # rendering. The max different pixels percentage is set based on the screen space taken up by the
 # env. The ``presets`` column selects a data-type variant on the consolidated cartpole camera task;
@@ -96,7 +101,10 @@ def test_rendering_registered_tasks(task_id: str, presets: str | None, env_name:
         from isaaclab_tasks.utils.parse_cfg import load_cfg_from_registry
 
         env_cfg = load_cfg_from_registry(task_id, "env_cfg_entry_point")
-        env_cfg = resolve_presets(env_cfg, {presets} if presets else frozenset())
+        selected = set(_PINNED_PRESETS)
+        if presets:
+            selected.add(presets)
+        env_cfg = resolve_presets(env_cfg, selected)
         env_cfg.sim.device = "cuda:0"
         env_cfg.scene.num_envs = 4
 
@@ -108,8 +116,8 @@ def test_rendering_registered_tasks(task_id: str, presets: str | None, env_name:
 
         maybe_save_stage(
             f"registered_tasks_{task_id}",
-            "default_physics",
-            "default_renderer",
+            _PINNED_PRESETS[0],
+            _PINNED_PRESETS[1],
             "stage",
             compare_golden=(presets is None),
         )
@@ -122,8 +130,8 @@ def test_rendering_registered_tasks(task_id: str, presets: str | None, env_name:
 
         validate_camera_outputs(
             f"registered_tasks/{task_id}",
-            "default_physics",
-            "default_renderer",
+            _PINNED_PRESETS[0],
+            _PINNED_PRESETS[1],
             camera_outputs,
             max_different_pixels_percentage=MAX_DIFFERENT_PIXELS_PERCENTAGE_BY_ENV_NAME[env_name],
             comparison_scores=_COMPARISON_SCORES,
diff --git a/source/isaaclab_tasks/test/core/test_runtime_compatibility.py b/source/isaaclab_tasks/test/core/test_runtime_compatibility.py
index 6ca07657def5..30acafbf7622 100644
--- a/source/isaaclab_tasks/test/core/test_runtime_compatibility.py
+++ b/source/isaaclab_tasks/test/core/test_runtime_compatibility.py
@@ -18,6 +18,7 @@
 import sys
 
 import pytest
+from isaaclab_newton.physics import NewtonCfg
 from isaaclab_ov.physics import OvPhysxCfg
 from isaaclab_ov.renderers import OVRTXRendererCfg
 from isaaclab_physx.physics import PhysxCfg
@@ -226,13 +227,14 @@ def test_newton_plus_ovrtx_is_valid():
     validate_runtime_compatibility(env_cfg)
 
 
-def test_default_isaacsim_physx_plus_ovrtx_raises():
-    """The concrete default Isaac Sim PhysX backend is incompatible with OVRTX."""
+def test_default_newton_plus_ovrtx_is_valid():
+    """The default Newton backend is compatible with OVRTX."""
     env_cfg = _resolve_with_presets("ovrtx")
 
-    assert isinstance(env_cfg.sim.physics, PhysxCfg)
-    with pytest.raises(ValueError, match="PhysxCfg"):
-        validate_runtime_compatibility(env_cfg)
+    assert isinstance(env_cfg.sim.physics, NewtonCfg)
+    config_scan = validate_runtime_compatibility(env_cfg)
+    assert isinstance(env_cfg.tiled_camera.renderer_cfg, OVRTXRendererCfg)
+    assert config_scan.needs_kit is False
 
 
 def test_explicit_auto_physx_plus_ovrtx_resolves_to_ovphysx():
@@ -302,14 +304,14 @@ def test_default_preset_is_valid():
     validate_runtime_compatibility(env_cfg)
 
 
-def test_rtx_with_default_physx_is_valid_and_resolves_to_isaac_sim_backends():
-    """The RTX selector follows the default concrete Isaac Sim PhysX backend."""
+def test_rtx_with_default_newton_is_valid_and_resolves_to_isaac_sim_backends():
+    """The RTX selector retains default Newton physics and requires Isaac Sim."""
     env_cfg = _resolve_with_presets("rtx")
     config_scan = validate_runtime_compatibility(env_cfg)
 
-    assert isinstance(env_cfg.sim.physics, PhysxCfg)
-    assert isinstance(env_cfg.tiled_camera.renderer_cfg, IsaacRtxRendererCfg)
-    assert config_scan.needs_kit is True
+    assert isinstance(env_cfg.sim.physics, NewtonCfg)
+    assert isinstance(env_cfg.tiled_camera.renderer_cfg, OVRTXRendererCfg)
+    assert config_scan.needs_kit is False
 
 
 def test_renderer_selector_physx_rtx_is_valid_and_resolves_to_ovphysx_and_ovrtx():
diff --git a/source/isaaclab_tasks/test/core/test_shadow_hand_camera_presets.py b/source/isaaclab_tasks/test/core/test_shadow_hand_camera_presets.py
index cad2801dab19..48ab3ce80b96 100644
--- a/source/isaaclab_tasks/test/core/test_shadow_hand_camera_presets.py
+++ b/source/isaaclab_tasks/test/core/test_shadow_hand_camera_presets.py
@@ -229,7 +229,7 @@ def test_all_camera_presets_present(shadow_hand_camera_presets):
 
 _RENDERER_PRESETS = [
     # preset_name, expected_class
-    ("default", IsaacRtxRendererCfg),
+    ("default", NewtonWarpRendererCfg),
     ("isaacsim_rtx", IsaacRtxRendererCfg),
     ("newton_renderer", NewtonWarpRendererCfg),
     pytest.param("ovrtx", OVRTXRendererCfg, marks=pytest.mark.skip(reason="OVRTX testing disabled")),
diff --git a/source/isaaclab_tasks/test/env_test_utils.py b/source/isaaclab_tasks/test/env_test_utils.py
index 3547c175fd66..c47fea95436a 100644
--- a/source/isaaclab_tasks/test/env_test_utils.py
+++ b/source/isaaclab_tasks/test/env_test_utils.py
@@ -5,6 +5,7 @@
 
 """Shared test utilities for Isaac Lab environments."""
 
+import gc
 import importlib
 import os
 import sys
@@ -521,6 +522,12 @@ def _check_random_actions(
         if env is not None:
             env.close()
 
+        # Drop unreachable environment objects while the device is still alive. Warp arrays
+        # free their device memory from a finalizer, so any array left for a later collection
+        # runs ``wp_free_device_async`` against streams that the teardown below has already
+        # destroyed, which segfaults the interpreter mid-run.
+        gc.collect()
+
         # Clear the simulation context singleton (also closes the USD context stage)
         SimulationContext.clear_instance()
 
diff --git a/source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Cartpole-Camera-Direct/default_physics-default_renderer-albedo.png b/source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Cartpole-Camera-Direct/isaacsim_physx-isaacsim_rtx-albedo.png
similarity index 100%
rename from source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Cartpole-Camera-Direct/default_physics-default_renderer-albedo.png
rename to source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Cartpole-Camera-Direct/isaacsim_physx-isaacsim_rtx-albedo.png
diff --git a/source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Cartpole-Camera-Direct/default_physics-default_renderer-depth.png b/source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Cartpole-Camera-Direct/isaacsim_physx-isaacsim_rtx-depth.png
similarity index 100%
rename from source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Cartpole-Camera-Direct/default_physics-default_renderer-depth.png
rename to source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Cartpole-Camera-Direct/isaacsim_physx-isaacsim_rtx-depth.png
diff --git a/source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Cartpole-Camera-Direct/default_physics-default_renderer-rgb.png b/source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Cartpole-Camera-Direct/isaacsim_physx-isaacsim_rtx-rgb.png
similarity index 100%
rename from source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Cartpole-Camera-Direct/default_physics-default_renderer-rgb.png
rename to source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Cartpole-Camera-Direct/isaacsim_physx-isaacsim_rtx-rgb.png
diff --git a/source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Cartpole-Camera-Direct/default_physics-default_renderer-rgba.png b/source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Cartpole-Camera-Direct/isaacsim_physx-isaacsim_rtx-rgba.png
similarity index 100%
rename from source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Cartpole-Camera-Direct/default_physics-default_renderer-rgba.png
rename to source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Cartpole-Camera-Direct/isaacsim_physx-isaacsim_rtx-rgba.png
diff --git a/source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Cartpole-Camera-Direct/default_physics-default_renderer-simple_shading_constant_diffuse.png b/source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Cartpole-Camera-Direct/isaacsim_physx-isaacsim_rtx-simple_shading_constant_diffuse.png
similarity index 100%
rename from source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Cartpole-Camera-Direct/default_physics-default_renderer-simple_shading_constant_diffuse.png
rename to source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Cartpole-Camera-Direct/isaacsim_physx-isaacsim_rtx-simple_shading_constant_diffuse.png
diff --git a/source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Cartpole-Camera-Direct/default_physics-default_renderer-simple_shading_diffuse_mdl.png b/source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Cartpole-Camera-Direct/isaacsim_physx-isaacsim_rtx-simple_shading_diffuse_mdl.png
similarity index 100%
rename from source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Cartpole-Camera-Direct/default_physics-default_renderer-simple_shading_diffuse_mdl.png
rename to source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Cartpole-Camera-Direct/isaacsim_physx-isaacsim_rtx-simple_shading_diffuse_mdl.png
diff --git a/source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Cartpole-Camera-Direct/default_physics-default_renderer-simple_shading_full_mdl.png b/source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Cartpole-Camera-Direct/isaacsim_physx-isaacsim_rtx-simple_shading_full_mdl.png
similarity index 100%
rename from source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Cartpole-Camera-Direct/default_physics-default_renderer-simple_shading_full_mdl.png
rename to source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Cartpole-Camera-Direct/isaacsim_physx-isaacsim_rtx-simple_shading_full_mdl.png
diff --git a/source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Repose-Cube-Shadow-Vision-Direct-v0/default_physics-default_renderer-depth.png b/source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Repose-Cube-Shadow-Vision-Direct-v0/isaacsim_physx-isaacsim_rtx-depth.png
similarity index 100%
rename from source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Repose-Cube-Shadow-Vision-Direct-v0/default_physics-default_renderer-depth.png
rename to source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Repose-Cube-Shadow-Vision-Direct-v0/isaacsim_physx-isaacsim_rtx-depth.png
diff --git a/source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Repose-Cube-Shadow-Vision-Direct-v0/default_physics-default_renderer-rgb.png b/source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Repose-Cube-Shadow-Vision-Direct-v0/isaacsim_physx-isaacsim_rtx-rgb.png
similarity index 100%
rename from source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Repose-Cube-Shadow-Vision-Direct-v0/default_physics-default_renderer-rgb.png
rename to source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Repose-Cube-Shadow-Vision-Direct-v0/isaacsim_physx-isaacsim_rtx-rgb.png
diff --git a/source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Repose-Cube-Shadow-Vision-Direct-v0/default_physics-default_renderer-rgba.png b/source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Repose-Cube-Shadow-Vision-Direct-v0/isaacsim_physx-isaacsim_rtx-rgba.png
similarity index 100%
rename from source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Repose-Cube-Shadow-Vision-Direct-v0/default_physics-default_renderer-rgba.png
rename to source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Repose-Cube-Shadow-Vision-Direct-v0/isaacsim_physx-isaacsim_rtx-rgba.png
diff --git a/source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Repose-Cube-Shadow-Vision-Direct-v0/default_physics-default_renderer-semantic_segmentation.png b/source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Repose-Cube-Shadow-Vision-Direct-v0/isaacsim_physx-isaacsim_rtx-semantic_segmentation.png
similarity index 100%
rename from source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Repose-Cube-Shadow-Vision-Direct-v0/default_physics-default_renderer-semantic_segmentation.png
rename to source/isaaclab_tasks/test/golden_images/registered_tasks/Isaac-Repose-Cube-Shadow-Vision-Direct-v0/isaacsim_physx-isaacsim_rtx-semantic_segmentation.png
diff --git a/source/isaaclab_tasks/test/golden_stages/registered_tasks_Isaac-Cartpole-Camera-Direct/default_physics-default_renderer-stage.usda b/source/isaaclab_tasks/test/golden_stages/registered_tasks_Isaac-Cartpole-Camera-Direct/isaacsim_physx-isaacsim_rtx-stage.usda
similarity index 100%
rename from source/isaaclab_tasks/test/golden_stages/registered_tasks_Isaac-Cartpole-Camera-Direct/default_physics-default_renderer-stage.usda
rename to source/isaaclab_tasks/test/golden_stages/registered_tasks_Isaac-Cartpole-Camera-Direct/isaacsim_physx-isaacsim_rtx-stage.usda
diff --git a/source/isaaclab_tasks/test/golden_stages/registered_tasks_Isaac-Reorient-Cube-Shadow-Camera-Direct/default_physics-default_renderer-stage.usda b/source/isaaclab_tasks/test/golden_stages/registered_tasks_Isaac-Reorient-Cube-Shadow-Camera-Direct/isaacsim_physx-isaacsim_rtx-stage.usda
similarity index 100%
rename from source/isaaclab_tasks/test/golden_stages/registered_tasks_Isaac-Reorient-Cube-Shadow-Camera-Direct/default_physics-default_renderer-stage.usda
rename to source/isaaclab_tasks/test/golden_stages/registered_tasks_Isaac-Reorient-Cube-Shadow-Camera-Direct/isaacsim_physx-isaacsim_rtx-stage.usda
diff --git a/source/isaaclab_visualizers/changelog.d/default-newton-rsl.skip b/source/isaaclab_visualizers/changelog.d/default-newton-rsl.skip
new file mode 100644
index 000000000000..e69de29bb2d1
diff --git a/source/isaaclab_visualizers/test/visualizer_integration_utils.py b/source/isaaclab_visualizers/test/visualizer_integration_utils.py
index 17c4351ad927..5b458ff574ad 100644
--- a/source/isaaclab_visualizers/test/visualizer_integration_utils.py
+++ b/source/isaaclab_visualizers/test/visualizer_integration_utils.py
@@ -1585,7 +1585,7 @@ def _make_anymal_d_env(visualizer_kind: str | tuple[str, ...], backend_kind: str
     from isaaclab.envs import ManagerBasedRLEnv
 
     env_cfg = copy.deepcopy(AnymalDFlatEnvCfg())
-    preset_key = "newton_mjwarp" if backend_kind == "newton" else "default"
+    preset_key = "newton_mjwarp" if backend_kind == "newton" else "isaacsim_physx"
     env_cfg.sim.physics = getattr(env_cfg.sim.physics, preset_key)
     env_cfg.scene.num_envs = (
         _ANYMAL_D_TILED_CAMERA_INTEGRATION_NUM_ENVS if tiled_camera else _ANYMAL_D_INTEGRATION_NUM_ENVS
@@ -1840,6 +1840,9 @@ def _make_cartpole_camera_env(
         env_cfg.observation_space = [th, tw, env_cfg.observation_space[2]]
     env_cfg.seed = None
     env_cfg.sim.physics, _ = _get_physics_cfg(backend_kind)
+    # Pin the camera renderer: these goldens compare visualizer output, so the frames must not
+    # shift when the MultiBackendRendererCfg default changes.
+    env_cfg.tiled_camera.default.renderer_cfg = env_cfg.tiled_camera.default.renderer_cfg.isaacsim_rtx
     visualizer_kinds = (visualizer_kind,) if isinstance(visualizer_kind, str) else tuple(visualizer_kind)
     visualizer_cfgs = [_get_visualizer_cfg(kind, tiled_camera=tiled_camera)[0] for kind in visualizer_kinds]
     env_cfg.sim.visualizer_cfgs = visualizer_cfgs[0] if len(visualizer_cfgs) == 1 else visualizer_cfgs