diff --git a/toolz/curried/__init__.py b/toolz/curried/__init__.py
index 43aeffd4..5cc01fac 100644
--- a/toolz/curried/__init__.py
+++ b/toolz/curried/__init__.py
@@ -82,6 +82,7 @@
 partitionby = toolz.curry(toolz.partitionby)
 pluck = toolz.curry(toolz.pluck)
 random_sample = toolz.curry(toolz.random_sample)
+randint_range = toolz.curry(toolz.randint_range)
 reduce = toolz.curry(toolz.reduce)
 reduceby = toolz.curry(toolz.reduceby)
 remove = toolz.curry(toolz.remove)
diff --git a/toolz/itertoolz.py b/toolz/itertoolz.py
index 2462c5fb..0a1e06a7 100644
--- a/toolz/itertoolz.py
+++ b/toolz/itertoolz.py
@@ -14,7 +14,8 @@
            'first', 'second', 'nth', 'last', 'get', 'concat', 'concatv',
            'mapcat', 'cons', 'interpose', 'frequencies', 'reduceby', 'iterate',
            'sliding_window', 'partition', 'partition_all', 'count', 'pluck',
-           'join', 'tail', 'diff', 'topk', 'peek', 'random_sample')
+           'join', 'tail', 'diff', 'topk', 'peek', 'random_sample',
+           'randint_range')
 
 
 def remove(predicate, seq):
@@ -979,3 +980,27 @@ def random_sample(prob, seq, random_state=None):
     if not hasattr(random_state, 'random'):
         random_state = Random(random_state)
     return filter(lambda _: random_state.random() < prob, seq)
+
+
+def randint_range(sample_size, min, max, random_state=None):
+    """ Generates a random sequence of integers from a specified range
+
+    Returns a lazy iterator of random integers from a range.
+
+    >>> list(randint_range(5, 100, 1000)) # doctest: +SKIP
+    [157, 196, 781, 882, 905]
+    >>> list(randint_range(5, 100, 1000)) # doctest: +SKIP
+    [234, 377, 501, 601, 885]
+    >>> list(randint_range(5, 100, 1000, random_state=2017))
+    [270, 548, 595, 878, 999]
+    >>> list(randint_range(5, 100, 1000, random_state=2017))
+    [270, 548, 595, 878, 999]
+    """
+    if not hasattr(random_state, 'random'):
+        random_state = Random(random_state)
+    population_size = max - min
+    for i in range(min, max):
+        if random_state.random() < sample_size / population_size:
+            yield i
+            sample_size -= 1
+        population_size -= 1