set OMP_NUM_THREADS

ml-agents/mlagents/trainers/learn.py

 )
 from mlagents.trainers.cli_utils import parser, DetectDefault
 from mlagents_envs.environment import UnityEnvironment
-from mlagents.trainers.settings import RunOptions
+from mlagents.trainers.settings import RunOptions, FrameworkType
 
 from mlagents.trainers.training_status import GlobalTrainingStatus
 from mlagents_envs.base_env import BaseEnv
             options.environment_parameters, run_seed, restore=checkpoint_settings.resume
         )
 
+        force_torch = "torch" in DetectDefault.non_default_args
+        if force_torch or any(
+            behavior.framework == FrameworkType.TENSORFLOW
+            for behavior in options.behaviors.values()
+        ):
+            os.environ["OMP_NUM_THREADS"] = "1"
+
         trainer_factory = TrainerFactory(
             trainer_config=options.behaviors,
             output_path=write_path,
             param_manager=env_parameter_manager,
             init_path=maybe_init_path,
             multi_gpu=False,
-            force_torch="torch" in DetectDefault.non_default_args,
+            force_torch=force_torch,
         )
         # Create controller and begin training.
         tc = TrainerController(