diff --git a/rllib/agents/trainer.py b/rllib/agents/trainer.py
index 92e252e52..b3a4de3cc 100644
--- a/rllib/agents/trainer.py
+++ b/rllib/agents/trainer.py
@@ -157,9 +157,6 @@ COMMON_CONFIG: TrainerConfigDict = {
     # makes it slightly harder to debug since Python code won't be evaluated
     # after the initial eager pass. Only possible if framework=tfe.
     "eager_tracing": False,
-    # Disable eager execution on workers (but allow it on the driver). This
-    # only has an effect if eager is enabled.
-    "no_eager_on_workers": False,
 
     # === Exploration Settings ===
     # Default exploration behavior, iff `explore`=None is passed into
diff --git a/rllib/evaluation/rollout_worker.py b/rllib/evaluation/rollout_worker.py
index 3097b6c22..899e2ef03 100644
--- a/rllib/evaluation/rollout_worker.py
+++ b/rllib/evaluation/rollout_worker.py
@@ -291,7 +291,6 @@ class RolloutWorker(ParallelIteratorWorker):
 
         policy_config: TrainerConfigDict = policy_config or {}
         if (tf1 and policy_config.get("framework") in ["tf2", "tfe"]
-                and not policy_config.get("no_eager_on_workers")
                 # This eager check is necessary for certain all-framework tests
                 # that use tf's eager_mode() context generator.
                 and not tf1.executing_eagerly()):