diff --git a/gemma_encoder.py b/gemma_encoder.py
index a60e7d6..c796a09 100644
--- a/gemma_encoder.py
+++ b/gemma_encoder.py
@@ -171,7 +171,8 @@ def __init__(self, embedding_directory=None, tokenizer_data={}):
 def ltxv_gemma_clip(encoder_path, ltxv_path, processor=None, dtype=None):
     class _LTXVGemmaTextEncoderModel(LTXVGemmaTextEncoderModel):
         def __init__(self, device="cpu", dtype=dtype, model_options={}):
-            dtype = torch.bfloat16  # TODO: make this configurable
+            if dtype is None:
+                dtype = torch.bfloat16
 
             gemma_model = Gemma3ForConditionalGeneration.from_pretrained(
                 encoder_path,
@@ -264,6 +265,15 @@ def load_model(self, gemma_path: str, ltxv_path: str, max_length: int):
             logger.warning(f"Could not load processor from {model_root}: {e}")
 
         clip_dtype = torch.bfloat16
+        try:
+            # MPS does not have full support bfloat16, use float16 instead
+            if comfy.model_management.get_torch_device().type == "mps":
+                clip_dtype = torch.float16
+        except Exception as e:
+            logger.debug(
+                f"Could not detect device type for dtype selection: {e}", exc_info=True
+            )
+
         ltxv_full_path = folder_paths.get_full_path("checkpoints", ltxv_path)
         clip_target = comfy.supported_models_base.ClipTarget(
             tokenizer=tokenizer_class,
@@ -428,9 +438,10 @@ def _enhance(
     )
     model_inputs = _pad_inputs_for_attention_alignment(model_inputs, pad_token_id)
 
+    devices = [model.device] if model.device.type == "cuda" else []
     with (
         torch.inference_mode(),
-        torch.random.fork_rng(devices=[model.device]),
+        torch.random.fork_rng(devices=devices),
         torch.autocast(device_type=model.device.type, dtype=model.dtype),
     ):
         torch.manual_seed(seed)