Missing inspect

unsloth/device_type.py

@@ -25,6 +25,7 @@ __all__ = [
 import torch
 import functools
 from unsloth_zoo.utils import Version
+import inspect
 
 @functools.cache
 def is_hip():

unsloth/models/loader.py

@@ -639,15 +639,14 @@ class FastModel(FastBaseModel):
                 "compatible with `full_finetuning=True`. If you wish to use QAT with LoRA, "
                 "please pass in `qat_scheme` in `FastLanguageModel.get_peft_model(...)` instead."
             )
-
-        old_model_name = model_name
-        if not use_exact_model_name:
-            model_name = get_model_name(model_name, load_in_4bit)
-
         # Check if 4bit is allowed specifically for AMD
         if not ALLOW_BITSANDBYTES:
             print("Unsloth: AMD currently is not stable with 4bit bitsandbytes. Disabling for now.")
             load_in_4bit = False
+
+        old_model_name = model_name
+        if not use_exact_model_name:
+            model_name = get_model_name(model_name, load_in_4bit)
         # Check if pre-quantized models are allowed
         # For eg AMD GPUs need blocksize = 128, but our pre-quants are blocksize = 64
         if not ALLOW_PREQUANTIZED_MODELS and model_name.endswith(("-unsloth-bnb-4bit", "-bnb-4bit")):