diff --git a/tensorflow_probability/python/distributions/__init__.py b/tensorflow_probability/python/distributions/__init__.py
index 6160c7d2f6..1c5998d5a0 100644
--- a/tensorflow_probability/python/distributions/__init__.py
+++ b/tensorflow_probability/python/distributions/__init__.py
@@ -65,6 +65,7 @@
 from tensorflow_probability.python.distributions.independent import Independent
 from tensorflow_probability.python.distributions.inflated import Inflated
 from tensorflow_probability.python.distributions.inflated import ZeroInflatedNegativeBinomial
+from tensorflow_probability.python.distributions.inflated import ZeroInflatedPoisson
 from tensorflow_probability.python.distributions.inverse_gamma import InverseGamma
 from tensorflow_probability.python.distributions.inverse_gaussian import InverseGaussian
 from tensorflow_probability.python.distributions.johnson_su import JohnsonSU
@@ -302,5 +303,6 @@
     'WishartLinearOperator',
     'WishartTriL',
     'ZeroInflatedNegativeBinomial',
+    'ZeroInflatedPoisson',
     'Zipf',
 ]
diff --git a/tensorflow_probability/python/distributions/inflated.py b/tensorflow_probability/python/distributions/inflated.py
index 0e69bee3ab..5a180c4ef6 100644
--- a/tensorflow_probability/python/distributions/inflated.py
+++ b/tensorflow_probability/python/distributions/inflated.py
@@ -25,6 +25,7 @@
 from tensorflow_probability.python.distributions import distribution as distribution_lib
 from tensorflow_probability.python.distributions import mixture
 from tensorflow_probability.python.distributions import negative_binomial
+from tensorflow_probability.python.distributions import poisson
 from tensorflow_probability.python.internal import auto_composite_tensor
 from tensorflow_probability.python.internal import dtype_util
 from tensorflow_probability.python.internal import parameter_properties
@@ -33,7 +34,8 @@
 from tensorflow_probability.python.internal import tensor_util
 from tensorflow_probability.python.util.deferred_tensor import DeferredTensor
 
-__all__ = ['Inflated', 'inflated_factory', 'ZeroInflatedNegativeBinomial']
+__all__ = ['Inflated', 'inflated_factory', 'ZeroInflatedNegativeBinomial',
+           'ZeroInflatedPoisson']
 
 
 def _safe_value_for_distribution(dist):
@@ -348,3 +350,8 @@ def another_property_getter(unused_self, value=v):
     negative_binomial.NegativeBinomial,
     0.0,
     require_integer_total_count=False)
+
+ZeroInflatedPoisson = inflated_factory(
+    'ZeroInflatedPoisson',
+    poisson.Poisson,
+    0.0)
diff --git a/tensorflow_probability/python/distributions/inflated_test.py b/tensorflow_probability/python/distributions/inflated_test.py
index 32baedbbc8..5cc8213d1b 100644
--- a/tensorflow_probability/python/distributions/inflated_test.py
+++ b/tensorflow_probability/python/distributions/inflated_test.py
@@ -21,6 +21,7 @@
 from tensorflow_probability.python.distributions import lognormal
 from tensorflow_probability.python.distributions import negative_binomial
 from tensorflow_probability.python.distributions import normal
+from tensorflow_probability.python.distributions import poisson
 from tensorflow_probability.python.experimental.util import trainable
 from tensorflow_probability.python.internal import test_util
 from tensorflow_probability.python.math import gradient
@@ -89,6 +90,36 @@ def test_zero_inflated_negative_binomial(self):
         inflated_loc_probs=0.2, probs=0.5, total_count=10.0)
     self.assertEqual('ZeroInflatedNegativeBinomial', zinb.name)
 
+  def test_zero_inflated_poisson(self):
+    zip_ = inflated.ZeroInflatedPoisson(
+        inflated_loc_probs=0.3, rate=5.0)
+    self.assertEqual('ZeroInflatedPoisson', zip_.name)
+    samples = zip_.sample(sample_shape=100, seed=test_util.test_seed())
+    self.assertEqual((100,), samples.shape)
+    lprob = zip_.log_prob(0.0)
+    self.assertAllFinite(lprob)
+    lprob = zip_.log_prob(3.0)
+    self.assertAllFinite(lprob)
+
+  def test_zero_inflated_poisson_logits(self):
+    zip_ = inflated.ZeroInflatedPoisson(
+        inflated_loc_logits=0.5, rate=3.0)
+    self.assertEqual('ZeroInflatedPoisson', zip_.name)
+    samples = zip_.sample(seed=test_util.test_seed())
+    self.assertAllFinite(samples)
+
+  def test_zero_inflated_poisson_batched(self):
+    poisson_dist = poisson.Poisson(
+        rate=np.array([1.0, 2.0, 3.0, 4.0, 5.0], dtype=np.float32))
+    zip_ = inflated.ZeroInflatedPoisson(
+        poisson_dist,
+        inflated_loc_probs=np.array([0.1, 0.2, 0.3, 0.4, 0.5],
+                                    dtype=np.float32))
+    lprob = zip_.log_prob([0, 1, 2, 3, 4])
+    self.assertEqual((5,), lprob.shape)
+    samples = zip_.sample(seed=test_util.test_seed())
+    self.assertEqual((5,), samples.shape)
+
   def test_zinb_is_trainable(self):
     init_fn, apply_fn = trainable.make_trainable_stateless(
         inflated.ZeroInflatedNegativeBinomial,