diff --git a/benchmarl/conf/experiment/base_experiment.yaml b/benchmarl/conf/experiment/base_experiment.yaml
index 6d371a58..4c0ccd34 100644
--- a/benchmarl/conf/experiment/base_experiment.yaml
+++ b/benchmarl/conf/experiment/base_experiment.yaml
@@ -56,6 +56,8 @@ off_policy_memory_size: 1_000_000
 off_policy_train_batch_size: 15_000
 
 evaluation: True
+# Whether to render the evaluation (if rendering is available)
+render: True
 # Frequency of evaluation in terms of experiment iterations
 evaluation_interval: 20
 # Number of episodes that evaluation is run on
diff --git a/benchmarl/experiment/experiment.py b/benchmarl/experiment/experiment.py
index e3cb95f8..83875c19 100644
--- a/benchmarl/experiment/experiment.py
+++ b/benchmarl/experiment/experiment.py
@@ -62,6 +62,7 @@ class ExperimentConfig:
     off_policy_train_batch_size: int = MISSING
 
     evaluation: bool = MISSING
+    render: bool = MISSING
     evaluation_interval: int = MISSING
     evaluation_episodes: int = MISSING
 
@@ -483,7 +484,7 @@ def _grad_clip(self, optimizer: torch.optim.Optimizer) -> float:
     def _evaluation_loop(self, iter: int):
         evaluation_start = time.time()
         with set_exploration_type(ExplorationType.MODE):
-            if self.task.has_render(self.test_env):
+            if self.task.has_render(self.test_env) and self.config.render:
                 frames = []
 
                 def callback(env, td):