diff --git a/docs/source/variational.rst b/docs/source/variational.rst
index 73354370a..cef92fd92 100644
--- a/docs/source/variational.rst
+++ b/docs/source/variational.rst
@@ -98,19 +98,23 @@ These are special :obj:`~gpytorch.variational._VariationalStrategy` objects that
 :obj:`~gpytorch.distributions.MultitaskMultivariateNormal` distributions. Each of these objects
 acts on a batch of approximate GPs.
 
-
-:hidden:`IndependentMultitaskVariationalStrategy`
+:hidden:`LMCVariationalStrategy`
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
-.. autoclass:: IndependentMultitaskVariationalStrategy
+.. autoclass:: LMCVariationalStrategy
    :members:
 
-:hidden:`LMCVariationalStrategy`
+   .. automethod:: __call__
+
+
+:hidden:`IndependentMultitaskVariationalStrategy`
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
-.. autoclass:: LMCVariationalStrategy
+.. autoclass:: IndependentMultitaskVariationalStrategy
    :members:
 
+   .. automethod:: __call__
+
 
 Variational Distributions
 -----------------------------
diff --git a/examples/04_Variational_and_Approximate_GPs/SVGP_Multitask_GP_Regression.ipynb b/examples/04_Variational_and_Approximate_GPs/SVGP_Multitask_GP_Regression.ipynb
index 1eabd15d5..fd5e2433c 100644
--- a/examples/04_Variational_and_Approximate_GPs/SVGP_Multitask_GP_Regression.ipynb
+++ b/examples/04_Variational_and_Approximate_GPs/SVGP_Multitask_GP_Regression.ipynb
@@ -176,7 +176,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -223,6 +223,64 @@
     "Note that all the batch sizes for `IndependentMultitaskVariationalStrategy` are now `num_tasks` rather than `num_latents`."
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Output modes\n",
+    "\n",
+    "By default, `LMCVariationalStrategy` and `IndependentMultitaskVariationalStrategy` produce vector-valued outputs. In other words, they return a `MultitaskMultivariateNormal` distribution -- containing all task values for each input.\n",
+    "\n",
+    "This is similar to the ExactGP model described in the [multitask GP regression tutorial](../03_Multitask_Exact_GPs/Multitask_GP_Regression.ipynb)."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "MultitaskMultivariateNormal torch.Size([100, 4])\n"
+     ]
+    }
+   ],
+   "source": [
+    "output = model(train_x)\n",
+    "print(output.__class__.__name__, output.event_shape)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Alternatively, if each input is only associated **with a single task**, passing in the `task_indices` argument will specify which task to return for each input. The result will be a standard `MultivariateNormal` distribution -- where each output corresponds to each input's specified task.\n",
+    "\n",
+    "This is similar to the ExactGP model described in the [Hadamard multitask GP regression tutorial](../03_Multitask_Exact_GPs/Hadamard_Multitask_GP_Regression.ipynb)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "MultivariateNormal torch.Size([5])\n"
+     ]
+    }
+   ],
+   "source": [
+    "x = train_x[..., :5]\n",
+    "task_indices = torch.LongTensor([0, 1, 3, 2, 2])\n",
+    "output = model(x, task_indices=task_indices)\n",
+    "print(output.__class__.__name__, output.event_shape)"
+   ]
+  },
   {
    "cell_type": "markdown",
    "metadata": {},
@@ -234,20 +292,28 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 8,
    "metadata": {
     "scrolled": false
    },
    "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/home/gpleiss/miniconda3/envs/gpytorch/lib/python3.7/site-packages/ipykernel_launcher.py:20: TqdmDeprecationWarning: This function will be removed in tqdm==5.0.0\n",
+      "Please use `tqdm.notebook.tqdm` instead of `tqdm.tqdm_notebook`\n"
+     ]
+    },
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "4b82e029261e40d58aafeeb640063467",
+       "model_id": "fffd43a1fd554e129b68581b060b0755",
        "version_major": 2,
        "version_minor": 0
       },
       "text/plain": [
-       "HBox(children=(IntProgress(value=0, description='Epoch', max=500, style=ProgressStyle(description_width='initi…"
+       "HBox(children=(HTML(value='Epoch'), FloatProgress(value=0.0, max=500.0), HTML(value='')))"
       ]
      },
      "metadata": {},
@@ -301,14 +367,14 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 9,
    "metadata": {
     "scrolled": true
    },
    "outputs": [
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 1152x216 with 4 Axes>"
       ]
diff --git a/gpytorch/variational/independent_multitask_variational_strategy.py b/gpytorch/variational/independent_multitask_variational_strategy.py
index 1c4edf861..9fbb461c3 100644
--- a/gpytorch/variational/independent_multitask_variational_strategy.py
+++ b/gpytorch/variational/independent_multitask_variational_strategy.py
@@ -2,7 +2,10 @@
 
 import warnings
 
-from ..distributions import MultitaskMultivariateNormal
+import torch
+
+from ..distributions import MultitaskMultivariateNormal, MultivariateNormal
+from ..lazy import RootLazyTensor
 from ..module import Module
 from ._variational_strategy import _VariationalStrategy
 
@@ -10,9 +13,12 @@
 class IndependentMultitaskVariationalStrategy(_VariationalStrategy):
     """
     IndependentMultitaskVariationalStrategy wraps an existing
-    :obj:`~gpytorch.variational.VariationalStrategy`
-    to produce a :obj:`~gpytorch.variational.MultitaskMultivariateNormal` distribution.
-    All outputs will be independent of one another.
+    :obj:`~gpytorch.variational.VariationalStrategy` to produce vector-valued (multi-task)
+    output distributions. Each task will be independent of one another.
+
+    The output will either be a :obj:`~gpytorch.distributions.MultitaskMultivariateNormal` distribution
+    (if we wish to evaluate all tasks for each input) or a :obj:`~gpytorch.distributions.MultivariateNormal`
+    (if we wish to evaluate a single task for each input).
 
     The base variational strategy is assumed to operate on a batch of GPs. One of the batch
     dimensions corresponds to the multiple tasks.
@@ -43,19 +49,46 @@ def variational_params_initialized(self):
     def kl_divergence(self):
         return super().kl_divergence().sum(dim=-1)
 
-    def __call__(self, x, prior=False, **kwargs):
+    def __call__(self, x, task_indices=None, prior=False, **kwargs):
+        r"""
+        See :class:`LMCVariationalStrategy`.
+        """
         function_dist = self.base_variational_strategy(x, prior=prior, **kwargs)
-        if (
-            self.task_dim > 0
-            and self.task_dim > len(function_dist.batch_shape)
-            or self.task_dim < 0
-            and self.task_dim + len(function_dist.batch_shape) < 0
-        ):
-            return MultitaskMultivariateNormal.from_repeated_mvn(function_dist, num_tasks=self.num_tasks)
+
+        if task_indices is None:
+            # Every data point will get an output for each task
+            if (
+                self.task_dim > 0
+                and self.task_dim > len(function_dist.batch_shape)
+                or self.task_dim < 0
+                and self.task_dim + len(function_dist.batch_shape) < 0
+            ):
+                return MultitaskMultivariateNormal.from_repeated_mvn(function_dist, num_tasks=self.num_tasks)
+            else:
+                function_dist = MultitaskMultivariateNormal.from_batch_mvn(function_dist, task_dim=self.task_dim)
+                assert function_dist.event_shape[-1] == self.num_tasks
+                return function_dist
+
         else:
-            function_dist = MultitaskMultivariateNormal.from_batch_mvn(function_dist, task_dim=self.task_dim)
-            assert function_dist.event_shape[-1] == self.num_tasks
-            return function_dist
+            # Each data point will get a single output corresponding to a single task
+
+            if self.task_dim > 0:
+                raise RuntimeError(f"task_dim must be a negative indexed batch dimension: got {self.task_dim}.")
+            num_batch = len(function_dist.batch_shape)
+            task_dim = num_batch + self.task_dim
+
+            # Create a mask to choose specific task assignment
+            shape = list(function_dist.batch_shape + function_dist.event_shape)
+            shape[task_dim] = 1
+            task_indices = task_indices.expand(shape).squeeze(task_dim)
+
+            # Create a mask to choose specific task assignment
+            task_mask = torch.nn.functional.one_hot(task_indices, num_classes=self.num_tasks)
+            task_mask = task_mask.permute(*range(0, task_dim), *range(task_dim + 1, num_batch + 1), task_dim)
+
+            mean = (function_dist.mean * task_mask).sum(task_dim)
+            covar = (function_dist.lazy_covariance_matrix * RootLazyTensor(task_mask[..., None])).sum(task_dim)
+            return MultivariateNormal(mean, covar)
 
 
 class MultitaskVariationalStrategy(IndependentMultitaskVariationalStrategy):
diff --git a/gpytorch/variational/lmc_variational_strategy.py b/gpytorch/variational/lmc_variational_strategy.py
index b3c43d25e..1e215affb 100644
--- a/gpytorch/variational/lmc_variational_strategy.py
+++ b/gpytorch/variational/lmc_variational_strategy.py
@@ -2,12 +2,34 @@
 
 import torch
 
-from ..distributions import MultitaskMultivariateNormal
-from ..lazy import KroneckerProductLazyTensor, MatmulLazyTensor
+from .. import settings
+from ..distributions import MultitaskMultivariateNormal, MultivariateNormal
+from ..lazy import KroneckerProductLazyTensor, RootLazyTensor
 from ..module import Module
+from ..utils.broadcasting import _mul_broadcast_shape
+from ..utils.interpolation import left_interp
 from ._variational_strategy import _VariationalStrategy
 
 
+def _select_lmc_coefficients(lmc_coefficients: torch.Tensor, indices: torch.LongTensor) -> torch.Tensor:
+    """
+    Given a list of indices for ... x N datapoints,
+      select the row from lmc_coefficient that corresponds to each datapoint
+
+    lmc_coefficients: torch.Tensor ... x num_latents x ... x num_tasks
+    indices: torch.Tesnor ... x N
+    """
+    batch_shape = _mul_broadcast_shape(lmc_coefficients.shape[:-1], indices.shape[:-1])
+
+    # We will use the left_interp helper to do the indexing
+    lmc_coefficients = lmc_coefficients.expand(*batch_shape, lmc_coefficients.shape[-1])[..., None]
+    indices = indices.expand(*batch_shape, indices.shape[-1])[..., None]
+    res = left_interp(
+        indices, torch.ones(indices.shape, dtype=torch.long, device=indices.device), lmc_coefficients,
+    ).squeeze(-1)
+    return res
+
+
 class LMCVariationalStrategy(_VariationalStrategy):
     r"""
     LMCVariationalStrategy is an implementation of the "Linear Model of Coregionalization"
@@ -20,8 +42,11 @@ class LMCVariationalStrategy(_VariationalStrategy):
 
         f_{\text{task } i}( \mathbf x) = \sum_{q=1}^Q a_i^{(q)} g^{(q)} ( \mathbf x )
 
-    LMCVariationalStrategy wraps an existing :obj:`~gpytorch.variational.VariationalStrategy`
-    to produce a :obj:`~gpytorch.variational.MultitaskMultivariateNormal` distribution.
+    LMCVariationalStrategy wraps an existing :obj:`~gpytorch.variational.VariationalStrategy`.
+    The output will either be a :obj:`~gpytorch.distributions.MultitaskMultivariateNormal` distribution
+    (if we wish to evaluate all tasks for each input) or a :obj:`~gpytorch.distributions.MultivariateNormal`
+    (if we wish to evaluate a single task for each input).
+
     The base variational strategy is assumed to operate on a multi-batch of GPs, where one
     of the batch dimensions corresponds to the latent function dimension.
 
@@ -35,13 +60,6 @@ class LMCVariationalStrategy(_VariationalStrategy):
         batch shape. This would correspond to each of the latent functions having different kernels
         or the same kernel, respectivly.
 
-    :param ~gpytorch.variational.VariationalStrategy base_variational_strategy: Base variational strategy
-    :param int num_tasks: The total number of tasks (output functions)
-    :param int num_latents: The total number of latent functions in each group
-    :param latent_dim: (Default: -1) Which batch dimension corresponds to the latent function batch.
-        **Must be negative indexed**
-    :type latent_dim: `int` < 0
-
     Example:
         >>> class LMCMultitaskGP(gpytorch.models.ApproximateGP):
         >>>     '''
@@ -74,7 +92,13 @@ class LMCVariationalStrategy(_VariationalStrategy):
         >>>             batch_shape=torch.Size([3]),
         >>>         )
         >>>
-        >>> # Model output: n x 5
+
+    :param ~gpytorch.variational.VariationalStrategy base_variational_strategy: Base variational strategy
+    :param int num_tasks: The total number of tasks (output functions)
+    :param int num_latents: The total number of latent functions in each group
+    :param latent_dim: (Default: -1) Which batch dimension corresponds to the latent function batch.
+        **Must be negative indexed**
+    :type latent_dim: `int` < 0
     """
 
     def __init__(
@@ -120,28 +144,84 @@ def variational_params_initialized(self):
     def kl_divergence(self):
         return super().kl_divergence().sum(dim=self.latent_dim)
 
-    def __call__(self, x, prior=False, **kwargs):
-        function_dist = self.base_variational_strategy(x, prior=prior, **kwargs)
-        lmc_coefficients = self.lmc_coefficients.expand(*function_dist.batch_shape, self.lmc_coefficients.size(-1))
-        num_batch = len(function_dist.batch_shape)
-        num_dim = num_batch + len(function_dist.event_shape)
-        latent_dim = num_batch + self.latent_dim if self.latent_dim is not None else None
-
-        # Mean
-        mean = function_dist.mean.permute(*range(0, latent_dim), *range(latent_dim + 1, num_dim), latent_dim)
-        mean = mean @ lmc_coefficients.permute(
-            *range(0, latent_dim), *range(latent_dim + 1, num_dim - 1), latent_dim, -1
-        )
-
-        # Covar
-        covar = function_dist.lazy_covariance_matrix
-        lmc_factor = MatmulLazyTensor(lmc_coefficients.unsqueeze(-1), lmc_coefficients.unsqueeze(-2))
-        covar = KroneckerProductLazyTensor(covar, lmc_factor)
-        covar = covar.sum(latent_dim)
-
-        # Add a bit of jitter to make the covar PD
-        covar = covar.add_jitter(1e-6)
-
-        # Done!
-        function_dist = MultitaskMultivariateNormal(mean, covar)
+    def __call__(self, x, task_indices=None, prior=False, **kwargs):
+        r"""
+        Computes the variational (or prior) distribution
+        :math:`q( \mathbf f \mid \mathbf X)` (or :math:`p( \mathbf f \mid \mathbf X)`).
+        There are two modes:
+
+        1.  Compute **all tasks** for all inputs.
+            If this is the case, the :attr:`task_indices` attribute should be None.
+            The return type will be a (... x N x num_tasks)
+            :class:`~gpytorch.distributions.MultitaskMultivariateNormal`.
+        2.  Compute **one task** per inputs.
+            If this is the case, the (... x N) :attr:`task_indices` tensor should contain
+            the indices of each input's assigned task.
+            The return type will be a (... x N)
+            :class:`~gpytorch.distributions.MultivariateNormal`.
+
+        :param x: Input locations to evaluate variational strategy
+        :type x: torch.Tensor (... x N x D)
+        :param task_indices: (Default: None) Task index associated with each input.
+            If this **is not** provided, then the returned distribution evaluates every input on every task
+            (returns :class:`~gpytorch.distributions.MultitaskMultivariateNormal`).
+            If this **is** provided, then the returned distribution evaluates each input only on its assigned task.
+            (returns :class:`~gpytorch.distributions.MultivariateNormal`).
+        :type task_indices: torch.Tensor (... x N), optional
+        :param prior: (Default: False) If False, returns the variational distribution
+            :math:`q( \mathbf f \mid \mathbf X)`.
+            If True, returns the prior distribution
+            :math:`p( \mathbf f \mid \mathbf X)`.
+        :type prior: bool
+        :return: :math:`q( \mathbf f \mid \mathbf X)` (or the prior),
+            either for all tasks (if `task_indices == None`)
+            or for a specific task (if `task_indices != None`).
+        :rtype: ~gpytorch.distributions.MultitaskMultivariateNormal (... x N x num_tasks)
+            or ~gpytorch.distributions.MultivariateNormal (... x N)
+        """
+        latent_dist = self.base_variational_strategy(x, prior=prior, **kwargs)
+        num_batch = len(latent_dist.batch_shape)
+        latent_dim = num_batch + self.latent_dim
+
+        if task_indices is None:
+            num_dim = num_batch + len(latent_dist.event_shape)
+
+            # Every data point will get an output for each task
+            # Therefore, we will set up the lmc_coefficients shape for a matmul
+            lmc_coefficients = self.lmc_coefficients.expand(*latent_dist.batch_shape, self.lmc_coefficients.size(-1))
+
+            # Mean: ... x N x num_tasks
+            latent_mean = latent_dist.mean.permute(*range(0, latent_dim), *range(latent_dim + 1, num_dim), latent_dim)
+            mean = latent_mean @ lmc_coefficients.permute(
+                *range(0, latent_dim), *range(latent_dim + 1, num_dim - 1), latent_dim, -1
+            )
+
+            # Covar: ... x (N x num_tasks) x (N x num_tasks)
+            latent_covar = latent_dist.lazy_covariance_matrix
+            lmc_factor = RootLazyTensor(lmc_coefficients.unsqueeze(-1))
+            covar = KroneckerProductLazyTensor(latent_covar, lmc_factor).sum(latent_dim)
+            # Add a bit of jitter to make the covar PD
+            covar = covar.add_jitter(settings.cholesky_jitter.value(dtype=mean.dtype))
+
+            # Done!
+            function_dist = MultitaskMultivariateNormal(mean, covar)
+
+        else:
+            # Each data point will get a single output corresponding to a single task
+            # Therefore, we will select the appropriate lmc coefficients for each task
+            lmc_coefficients = _select_lmc_coefficients(self.lmc_coefficients, task_indices)
+
+            # Mean: ... x N
+            mean = (latent_dist.mean * lmc_coefficients).sum(latent_dim)
+
+            # Covar: ... x N x N
+            latent_covar = latent_dist.lazy_covariance_matrix
+            lmc_factor = RootLazyTensor(lmc_coefficients.unsqueeze(-1))
+            covar = (latent_covar * lmc_factor).sum(latent_dim)
+            # Add a bit of jitter to make the covar PD
+            covar = covar.add_jitter(settings.cholesky_jitter.value(dtype=mean.dtype))
+
+            # Done!
+            function_dist = MultivariateNormal(mean, covar)
+
         return function_dist
diff --git a/test/examples/test_lmc_svgp_regression.py b/test/examples/test_lmc_svgp_regression.py
index c2958e3bc..822bbecb0 100644
--- a/test/examples/test_lmc_svgp_regression.py
+++ b/test/examples/test_lmc_svgp_regression.py
@@ -7,7 +7,7 @@
 
 import gpytorch
 import torch
-from gpytorch.likelihoods import MultitaskGaussianLikelihood
+from gpytorch.likelihoods import GaussianLikelihood, MultitaskGaussianLikelihood
 
 
 # Batch training test: Let's learn hyperparameters on a sine dataset, but test on a sine dataset and a cosine dataset
@@ -75,7 +75,6 @@ def tearDown(self):
             torch.set_rng_state(self.rng_state)
 
     def test_train_and_eval(self):
-        # We're manually going to set the hyperparameters to something they shouldn't be
         likelihood = MultitaskGaussianLikelihood(num_tasks=4)
         model = LMCModel()
 
@@ -132,6 +131,57 @@ def test_train_and_eval(self):
             self.assertEqual(lower.shape, train_y.shape)
             self.assertEqual(upper.shape, train_y.shape)
 
+    def test_indexed_train_and_eval(self):
+        likelihood = GaussianLikelihood()
+        model = LMCModel()
+
+        # Find optimal model hyperparameters
+        model.train()
+        likelihood.train()
+        optimizer = torch.optim.Adam([
+            {'params': model.parameters()},
+            {'params': likelihood.parameters()},
+        ], lr=0.01)
+
+        # Our loss object. We're using the VariationalELBO, which essentially just computes the ELBO
+        mll = gpytorch.mlls.VariationalELBO(likelihood, model, num_data=train_y.size(0))
+
+        # Create some task indices
+        arange = torch.arange(train_x.size(0))
+        train_i = torch.rand(train_x.size(0)).mul(4).floor().long()
+
+        # We use more CG iterations here because the preconditioner introduced in the NeurIPS paper seems to be less
+        # effective for VI.
+        for i in range(400):
+            # Within each iteration, we will go over each minibatch of data
+            optimizer.zero_grad()
+            output = model(train_x, task_indices=train_i)
+            loss = -mll(output, train_y[arange, train_i])
+            loss.backward()
+            optimizer.step()
+
+            for param in model.parameters():
+                self.assertTrue(param.grad is not None)
+                self.assertGreater(param.grad.norm().item(), 0)
+            for param in likelihood.parameters():
+                self.assertTrue(param.grad is not None)
+                self.assertGreater(param.grad.norm().item(), 0)
+
+        # Test the model
+        model.eval()
+        likelihood.eval()
+
+        # Make predictions for both sets of test points, and check MAEs.
+        with torch.no_grad(), gpytorch.settings.max_eager_kernel_size(1):
+            predictions = likelihood(model(train_x, task_indices=train_i))
+            mean_abs_error = torch.mean(torch.abs(train_y[arange, train_i] - predictions.mean))
+            self.assertLess(mean_abs_error.squeeze().item(), 0.15)
+
+            # Smoke test for getting predictive uncertainties
+            lower, upper = predictions.confidence_region()
+            self.assertEqual(lower.shape, train_i.shape)
+            self.assertEqual(upper.shape, train_i.shape)
+
 
 if __name__ == "__main__":
     unittest.main()
diff --git a/test/variational/test_independent_multitask_variational_strategy.py b/test/variational/test_independent_multitask_variational_strategy.py
index c04e6e018..ab88f52ed 100644
--- a/test/variational/test_independent_multitask_variational_strategy.py
+++ b/test/variational/test_independent_multitask_variational_strategy.py
@@ -8,10 +8,14 @@
 from gpytorch.test.variational_test_case import VariationalTestCase
 
 
-def likelihood_cls():
+def multitask_likelihood_cls():
     return gpytorch.likelihoods.MultitaskGaussianLikelihood(num_tasks=2)
 
 
+def singletask_likelihood_cls():
+    return gpytorch.likelihoods.GaussianLikelihood()
+
+
 def strategy_cls(model, inducing_points, variational_distribution, learn_inducing_locations):
     return gpytorch.variational.IndependentMultitaskVariationalStrategy(
         gpytorch.variational.VariationalStrategy(
@@ -36,7 +40,7 @@ def distribution_cls(self):
 
     @property
     def likelihood_cls(self):
-        return likelihood_cls
+        return multitask_likelihood_cls
 
     @property
     def mll_cls(self):
@@ -49,22 +53,12 @@ def strategy_cls(self):
     def test_training_iteration(self, *args, expected_batch_shape=None, **kwargs):
         expected_batch_shape = expected_batch_shape or self.batch_shape
         expected_batch_shape = expected_batch_shape[:-1]
-        cg_mock, cholesky_mock, ciq_mock = super().test_training_iteration(
-            *args, expected_batch_shape=expected_batch_shape, **kwargs
-        )
-        self.assertFalse(cg_mock.called)
-        self.assertFalse(ciq_mock.called)
-        self.assertEqual(cholesky_mock.call_count, 2)  # One for each forward pass
+        super().test_training_iteration(*args, expected_batch_shape=expected_batch_shape, **kwargs)
 
     def test_eval_iteration(self, *args, expected_batch_shape=None, **kwargs):
         expected_batch_shape = expected_batch_shape or self.batch_shape
         expected_batch_shape = expected_batch_shape[:-1]
-        cg_mock, cholesky_mock, ciq_mock = super().test_eval_iteration(
-            *args, expected_batch_shape=expected_batch_shape, **kwargs
-        )
-        self.assertFalse(cg_mock.called)
-        self.assertFalse(ciq_mock.called)
-        self.assertEqual(cholesky_mock.call_count, 1)  # One to compute cache, that's it!
+        super().test_eval_iteration(*args, expected_batch_shape=expected_batch_shape, **kwargs)
 
 
 class TestMultitaskPredictiveGP(TestMultitaskVariationalGP):
@@ -115,5 +109,115 @@ def distribution_cls(self):
         return gpytorch.variational.DeltaVariationalDistribution
 
 
+class TestIndexedMultitaskVariationalGP(TestMultitaskVariationalGP, unittest.TestCase):
+    def _training_iter(
+        self, model, likelihood, batch_shape=torch.Size([]), mll_cls=gpytorch.mlls.VariationalELBO, cuda=False
+    ):
+        batch_shape = list(batch_shape)
+        batch_shape[-1] = 1
+        train_x = torch.randn(*batch_shape, 32, 2).clamp(-2.5, 2.5)
+        train_i = torch.rand(*batch_shape, 32).round().long()
+        train_y = torch.linspace(-1, 1, self.event_shape[0])
+        train_y = train_y.view(self.event_shape[0], *([1] * (len(self.event_shape) - 1)))
+        train_y = train_y.expand(*self.event_shape)
+        mll = mll_cls(likelihood, model, num_data=train_x.size(-2))
+        if cuda:
+            train_x = train_x.cuda()
+            train_i = train_i.cuda()
+            train_y = train_y.cuda()
+            model = model.cuda()
+            likelihood = likelihood.cuda()
+
+        # Single optimization iteration
+        model.train()
+        likelihood.train()
+        output = model(train_x, task_indices=train_i)
+        loss = -mll(output, train_y)
+        loss.sum().backward()
+
+        # Make sure we have gradients for all parameters
+        for _, param in model.named_parameters():
+            self.assertTrue(param.grad is not None)
+            self.assertGreater(param.grad.norm().item(), 0)
+        for _, param in likelihood.named_parameters():
+            self.assertTrue(param.grad is not None)
+            self.assertGreater(param.grad.norm().item(), 0)
+
+        return output, loss
+
+    def _eval_iter(self, model, batch_shape=torch.Size([]), cuda=False):
+        batch_shape = list(batch_shape)
+        batch_shape[-1] = 1
+        test_x = torch.randn(*batch_shape, 32, 2).clamp(-2.5, 2.5)
+        test_i = torch.rand(*batch_shape, 32).round().long()
+        if cuda:
+            test_x = test_x.cuda()
+            test_i = test_i.cuda()
+            model = model.cuda()
+
+        # Single optimization iteration
+        model.eval()
+        with torch.no_grad():
+            output = model(test_x, task_indices=test_i)
+
+        return output
+
+    @property
+    def event_shape(self):
+        return torch.Size([32])
+
+    @property
+    def likelihood_cls(self):
+        return singletask_likelihood_cls
+
+
+class TestIndexedMultitaskPredictiveGP(TestIndexedMultitaskVariationalGP):
+    @property
+    def mll_cls(self):
+        return gpytorch.mlls.PredictiveLogLikelihood
+
+
+class TestIndexedMultitaskRobustVGP(TestIndexedMultitaskVariationalGP):
+    @property
+    def mll_cls(self):
+        return gpytorch.mlls.GammaRobustVariationalELBO
+
+
+class TestMeanFieldIndexedMultitaskVariationalGP(TestIndexedMultitaskVariationalGP):
+    @property
+    def distribution_cls(self):
+        return gpytorch.variational.MeanFieldVariationalDistribution
+
+
+class TestMeanFieldIndexedMultitaskPredictiveGP(TestIndexedMultitaskPredictiveGP):
+    @property
+    def distribution_cls(self):
+        return gpytorch.variational.MeanFieldVariationalDistribution
+
+
+class TestMeanFieldIndexedMultitaskRobustVGP(TestIndexedMultitaskRobustVGP):
+    @property
+    def distribution_cls(self):
+        return gpytorch.variational.MeanFieldVariationalDistribution
+
+
+class TestDeltaIndexedMultitaskVariationalGP(TestIndexedMultitaskVariationalGP):
+    @property
+    def distribution_cls(self):
+        return gpytorch.variational.DeltaVariationalDistribution
+
+
+class TestDeltaIndexedMultitaskPredictiveGP(TestIndexedMultitaskPredictiveGP):
+    @property
+    def distribution_cls(self):
+        return gpytorch.variational.DeltaVariationalDistribution
+
+
+class TestDeltaIndexedMultitaskRobustVGP(TestIndexedMultitaskRobustVGP):
+    @property
+    def distribution_cls(self):
+        return gpytorch.variational.DeltaVariationalDistribution
+
+
 if __name__ == "__main__":
     unittest.main()
diff --git a/test/variational/test_lmc_variational_strategy.py b/test/variational/test_lmc_variational_strategy.py
index e1d9ad10a..2e5255200 100644
--- a/test/variational/test_lmc_variational_strategy.py
+++ b/test/variational/test_lmc_variational_strategy.py
@@ -8,10 +8,14 @@
 from gpytorch.test.variational_test_case import VariationalTestCase
 
 
-def likelihood_cls():
+def multitask_likelihood_cls():
     return gpytorch.likelihoods.MultitaskGaussianLikelihood(num_tasks=4)
 
 
+def singletask_likelihood_cls():
+    return gpytorch.likelihoods.GaussianLikelihood()
+
+
 def strategy_cls(model, inducing_points, variational_distribution, learn_inducing_locations):
     return gpytorch.variational.LMCVariationalStrategy(
         gpytorch.variational.VariationalStrategy(
@@ -38,7 +42,7 @@ def distribution_cls(self):
 
     @property
     def likelihood_cls(self):
-        return likelihood_cls
+        return multitask_likelihood_cls
 
     @property
     def mll_cls(self):
@@ -167,5 +171,111 @@ def distribution_cls(self):
         return gpytorch.variational.DeltaVariationalDistribution
 
 
+class TestIndexedLMCVariationalGP(TestLMCVariationalGP, unittest.TestCase):
+    def _training_iter(
+        self, model, likelihood, batch_shape=torch.Size([]), mll_cls=gpytorch.mlls.VariationalELBO, cuda=False
+    ):
+        train_x = torch.randn(*batch_shape, 32, 2).clamp(-2.5, 2.5)
+        train_i = torch.rand(*batch_shape, 32).round().long()
+        train_y = torch.linspace(-1, 1, self.event_shape[0])
+        train_y = train_y.view(self.event_shape[0], *([1] * (len(self.event_shape) - 1)))
+        train_y = train_y.expand(*self.event_shape)
+        mll = mll_cls(likelihood, model, num_data=train_x.size(-2))
+        if cuda:
+            train_x = train_x.cuda()
+            train_i = train_i.cuda()
+            train_y = train_y.cuda()
+            model = model.cuda()
+            likelihood = likelihood.cuda()
+
+        # Single optimization iteration
+        model.train()
+        likelihood.train()
+        output = model(train_x, task_indices=train_i)
+        loss = -mll(output, train_y)
+        loss.sum().backward()
+
+        # Make sure we have gradients for all parameters
+        for _, param in model.named_parameters():
+            self.assertTrue(param.grad is not None)
+            self.assertGreater(param.grad.norm().item(), 0)
+        for _, param in likelihood.named_parameters():
+            self.assertTrue(param.grad is not None)
+            self.assertGreater(param.grad.norm().item(), 0)
+
+        return output, loss
+
+    def _eval_iter(self, model, batch_shape=torch.Size([]), cuda=False):
+        test_x = torch.randn(*batch_shape, 32, 2).clamp(-2.5, 2.5)
+        test_i = torch.rand(*batch_shape, 32).round().long()
+        if cuda:
+            test_x = test_x.cuda()
+            test_i = test_i.cuda()
+            model = model.cuda()
+
+        # Single optimization iteration
+        model.eval()
+        with torch.no_grad():
+            output = model(test_x, task_indices=test_i)
+
+        return output
+
+    @property
+    def event_shape(self):
+        return torch.Size([32])
+
+    @property
+    def likelihood_cls(self):
+        return singletask_likelihood_cls
+
+
+class TestIndexedLMCPredictiveGP(TestIndexedLMCVariationalGP):
+    @property
+    def mll_cls(self):
+        return gpytorch.mlls.PredictiveLogLikelihood
+
+
+class TestIndexedLMCRobustVGP(TestIndexedLMCVariationalGP):
+    @property
+    def mll_cls(self):
+        return gpytorch.mlls.GammaRobustVariationalELBO
+
+
+class TestMeanFieldIndexedLMCVariationalGP(TestIndexedLMCVariationalGP):
+    @property
+    def distribution_cls(self):
+        return gpytorch.variational.MeanFieldVariationalDistribution
+
+
+class TestMeanFieldIndexedLMCPredictiveGP(TestIndexedLMCPredictiveGP):
+    @property
+    def distribution_cls(self):
+        return gpytorch.variational.MeanFieldVariationalDistribution
+
+
+class TestMeanFieldIndexedLMCRobustVGP(TestIndexedLMCRobustVGP):
+    @property
+    def distribution_cls(self):
+        return gpytorch.variational.MeanFieldVariationalDistribution
+
+
+class TestDeltaIndexedLMCVariationalGP(TestIndexedLMCVariationalGP):
+    @property
+    def distribution_cls(self):
+        return gpytorch.variational.DeltaVariationalDistribution
+
+
+class TestDeltaIndexedLMCPredictiveGP(TestIndexedLMCPredictiveGP):
+    @property
+    def distribution_cls(self):
+        return gpytorch.variational.DeltaVariationalDistribution
+
+
+class TestDeltaIndexedLMCRobustVGP(TestIndexedLMCRobustVGP):
+    @property
+    def distribution_cls(self):
+        return gpytorch.variational.DeltaVariationalDistribution
+
+
 if __name__ == "__main__":
     unittest.main()