diff --git a/src/autotrain/cli/run_llm.py b/src/autotrain/cli/run_llm.py
index 53bb2cb985..975fd13516 100644
--- a/src/autotrain/cli/run_llm.py
+++ b/src/autotrain/cli/run_llm.py
@@ -478,7 +478,7 @@ def __init__(self, args):
             self.num_gpus = 1
 
     def run(self):
-        from autotrain.backend import EndpointsRunner, SpaceRunner
+        from autotrain.backend import SpaceRunner
         from autotrain.trainers.clm.__main__ import train as train_llm
         from autotrain.trainers.clm.params import LLMTrainingParams
 
@@ -544,16 +544,6 @@ def run(self):
                 logger.info(f"Training Space created. Check progress at https://hf.co/spaces/{space_id}")
                 sys.exit(0)
 
-            if self.args.backend.startswith("ep-"):
-                logger.info("Creating training endpoint...")
-                sr = EndpointsRunner(
-                    params=params,
-                    backend=self.args.backend,
-                )
-                sr.prepare()
-                logger.info("Training endpoint created.")
-                sys.exit(0)
-
             # local training
             params.save(output_dir=self.args.project_name)
             if self.num_gpus == 1:
diff --git a/src/autotrain/cli/run_seq2seq.py b/src/autotrain/cli/run_seq2seq.py
index e961589acd..697ca87000 100644
--- a/src/autotrain/cli/run_seq2seq.py
+++ b/src/autotrain/cli/run_seq2seq.py
@@ -6,7 +6,7 @@
 import torch
 
 from autotrain import logger
-from autotrain.backend import EndpointsRunner, SpaceRunner
+from autotrain.backend import SpaceRunner
 
 from . import BaseAutoTrainCommand
 
@@ -397,16 +397,6 @@ def run(self):
                 logger.info(f"Training Space created. Check progress at https://hf.co/spaces/{space_id}")
                 sys.exit(0)
 
-            if self.args.backend.startswith("ep-"):
-                logger.info("Creating training endpoint...")
-                sr = EndpointsRunner(
-                    params=params,
-                    backend=self.args.backend,
-                )
-                sr.prepare()
-                logger.info("Training endpoint created.")
-                sys.exit(0)
-
             params.save(output_dir=self.args.project_name)
             if self.num_gpus == 1:
                 train_seq2seq(params)
diff --git a/src/autotrain/cli/run_tabular.py b/src/autotrain/cli/run_tabular.py
index e724d29495..b1151bb3dd 100644
--- a/src/autotrain/cli/run_tabular.py
+++ b/src/autotrain/cli/run_tabular.py
@@ -5,7 +5,7 @@
 import torch
 
 from autotrain import logger
-from autotrain.backend import EndpointsRunner, SpaceRunner
+from autotrain.backend import SpaceRunner
 
 from . import BaseAutoTrainCommand
 
@@ -265,15 +265,5 @@ def run(self):
                 logger.info(f"Training Space created. Check progress at https://hf.co/spaces/{space_id}")
                 sys.exit(0)
 
-            if self.args.backend.startswith("ep-"):
-                logger.info("Creating training endpoint...")
-                sr = EndpointsRunner(
-                    params=params,
-                    backend=self.args.backend,
-                )
-                sr.prepare()
-                logger.info("Training endpoint created.")
-                sys.exit(0)
-
             params.save(output_dir=self.args.project_name)
             train_tabular(params)
diff --git a/src/autotrain/cli/run_text_classification.py b/src/autotrain/cli/run_text_classification.py
index e700c772af..82e1f03f29 100644
--- a/src/autotrain/cli/run_text_classification.py
+++ b/src/autotrain/cli/run_text_classification.py
@@ -6,7 +6,7 @@
 import torch
 
 from autotrain import logger
-from autotrain.backend import EndpointsRunner, SpaceRunner
+from autotrain.backend import SpaceRunner
 
 from . import BaseAutoTrainCommand
 
@@ -346,16 +346,6 @@ def run(self):
                 logger.info(f"Training Space created. Check progress at https://hf.co/spaces/{space_id}")
                 sys.exit(0)
 
-            if self.args.backend.startswith("ep-"):
-                logger.info("Creating training endpoint...")
-                sr = EndpointsRunner(
-                    params=params,
-                    backend=self.args.backend,
-                )
-                sr.prepare()
-                logger.info("Training endpoint created.")
-                sys.exit(0)
-
             params.save(output_dir=self.args.project_name)
             if self.num_gpus == 1:
                 train_text_classification(params)