diff --git a/src/vllm_tgis_adapter/tgis_utils/args.py b/src/vllm_tgis_adapter/tgis_utils/args.py
index 7be4ce1..475e14a 100644
--- a/src/vllm_tgis_adapter/tgis_utils/args.py
+++ b/src/vllm_tgis_adapter/tgis_utils/args.py
@@ -15,6 +15,10 @@ def _to_env_var(arg_name: str) -> str:
     return arg_name.upper().replace("-", "_")
 
 
+def _bool_from_string(val: str) -> bool:
+    return val.lower().strip() == "true" or val == "1"
+
+
 def _switch_action_default(action: argparse.Action) -> None:
     """Switch to using env var fallback for all args."""
     env_val = os.environ.get(_to_env_var(action.dest))
@@ -22,8 +26,8 @@ def _switch_action_default(action: argparse.Action) -> None:
         return
 
     val: bool | str | int
-    if action.type is bool:
-        val = env_val.lower() == "true" or env_val == "1"
+    if action.type in [bool, _bool_from_string]:
+        val = _bool_from_string(env_val)
     elif action.type is int:
         val = int(env_val)
     else:
@@ -111,9 +115,11 @@ def add_tgis_args(parser: argparse.ArgumentParser) -> argparse.ArgumentParser:
     # map to tensor_parallel_size
     parser.add_argument("--num-shard", type=int)
     # TODO check boolean behaviour for env vars and defaults
-    parser.add_argument("--output-special-tokens", type=bool, default=False)
     parser.add_argument(
-        "--default-include-stop-seqs", type=bool, default=True
+        "--output-special-tokens", type=_bool_from_string, default=False
+    )
+    parser.add_argument(
+        "--default-include-stop-seqs", type=_bool_from_string, default=True
     )  # TODO TBD
     parser.add_argument("--grpc-port", type=int, default=8033)
 
@@ -135,9 +141,13 @@ def add_tgis_args(parser: argparse.ArgumentParser) -> argparse.ArgumentParser:
     parser.add_argument("--speculator-n-candidates", type=int)
     parser.add_argument("--speculator-max-batch-size", type=int)
     # allow re-enabling vllm native per-request logging
-    parser.add_argument("--enable-vllm-log-requests", type=bool, default=False)
+    parser.add_argument(
+        "--enable-vllm-log-requests", type=_bool_from_string, default=False
+    )
     # set to true to disable producing prompt logprobs on all requests
-    parser.add_argument("--disable-prompt-logprobs", type=bool, default=False)
+    parser.add_argument(
+        "--disable-prompt-logprobs", type=_bool_from_string, default=False
+    )
 
     # TODO check/add other args here