diff --git a/unsloth/models/vision.py b/unsloth/models/vision.py
index e897145acc..cb8909818d 100644
--- a/unsloth/models/vision.py
+++ b/unsloth/models/vision.py
@@ -525,7 +525,7 @@ class FastBaseModel:
                         f"To enable bfloat16 training to reduce VRAM usage by 50% albeit with a slightly higher loss, "\
                         "use `float32_mixed_precision = False` during FastLanguageModel.from_pretrained"
                     )
-                    os.environ["UNSLOTH_BFLOAT16_MIXED_PRECISION"] = "1"
+                os.environ["UNSLOTH_BFLOAT16_MIXED_PRECISION"] = "1"
             else:
                 print("Unsloth: Float16 full finetuning uses more memory since we upcast weights to float32.")
         else: