diff --git a/fairseq/optim/adadelta.py b/fairseq/optim/adadelta.py
new file mode 100644
index 000000000..f9d82047c
--- /dev/null
+++ b/fairseq/optim/adadelta.py
@@ -0,0 +1,40 @@
+# Copyright (c) 2017-present, Facebook, Inc.
+# All rights reserved.
+#
+# This source code is licensed under the license found in the LICENSE file in
+# the root directory of this source tree. An additional grant of patent rights
+# can be found in the PATENTS file in the same directory.
+
+import torch.optim
+
+from . import FairseqOptimizer, register_optimizer
+
+
+@register_optimizer('adadelta')
+class Adadelta(FairseqOptimizer):
+    def __init__(self, args, params):
+        super().__init__(args, params)
+        self._optimizer = torch.optim.Adadelta(params, **self.optimizer_config)
+
+    @staticmethod
+    def add_args(parser):
+        """Add optimizer-specific arguments to the parser."""
+        parser.add_argument('--adadelta-rho', type=float, default=0.9, metavar='RHO',
+                            help='coefficient used for computing a running average of squared gradients')
+        parser.add_argument('--adadelta-eps', type=float, default=1e-6, metavar='EPS',
+                            help='term added to the denominator to improve numerical stability')
+
+    @property
+    def optimizer_config(self):
+        """
+        Return a kwarg dictionary that will be used to override optimizer
+        args stored in checkpoints. This allows us to load a checkpoint and
+        resume training using a different set of optimizer args, e.g., with a
+        different learning rate.
+        """
+        return {
+            'lr': self.args.lr[0],
+            'rho': self.args.adadelta_rho,
+            'eps': self.args.adadelta_eps,
+            'weight_decay': self.args.weight_decay,
+        }
diff --git a/tests/test_binaries.py b/tests/test_binaries.py
index feff2f282..69276e498 100644
--- a/tests/test_binaries.py
+++ b/tests/test_binaries.py
@@ -220,6 +220,28 @@ class TestLanguageModeling(unittest.TestCase):
                 eval_lm_main(data_dir)
 
 
+class TestCommonOptions(unittest.TestCase):
+
+    def test_optimizers(self):
+        with contextlib.redirect_stdout(StringIO()):
+            with tempfile.TemporaryDirectory('test_optimizers') as data_dir:
+                # Use just a bit of data and tiny model to keep this test runtime reasonable
+                create_dummy_data(data_dir, num_examples=10, maxlen=5)
+                preprocess_translation_data(data_dir)
+                optimizers = ['adafactor', 'adam', 'nag', 'adagrad', 'sgd', 'adadelta']
+                last_checkpoint = os.path.join(data_dir, 'checkpoint_last.pt')
+                for optimizer in optimizers:
+                    if os.path.exists(last_checkpoint):
+                        os.remove(last_checkpoint)
+                    train_translation_model(data_dir, 'lstm', [
+                        '--encoder-layers', '1',
+                        '--encoder-hidden-size', '32',
+                        '--decoder-layers', '1',
+                        '--optimizer', optimizer,
+                    ])
+                    generate_main(data_dir)
+
+
 def create_dummy_data(data_dir, num_examples=1000, maxlen=20):
 
     def _create_dummy_data(filename):
@@ -267,7 +289,6 @@ def train_translation_model(data_dir, arch, extra_flags=None):
             data_dir,
             '--save-dir', data_dir,
             '--arch', arch,
-            '--optimizer', 'nag',
             '--lr', '0.05',
             '--max-tokens', '500',
             '--max-epoch', '1',