Quant Method missing

Author: danielhanchen

unsloth/__init__.py

@@ -89,8 +89,10 @@
         #         except:
         #             raise ImportError("Unsloth: Please update unsloth_zoo via `pip install --upgrade --no-cache-dir --no-deps unsloth_zoo`")
     import unsloth_zoo
-except:
-    raise ImportError("Unsloth: Please install unsloth_zoo via `pip install unsloth_zoo`")
+except NotImplementedError as e:
+    raise NotImplementedError(str(e))
+except Exception as e:
+    raise ImportError(f"Unsloth: Please install unsloth_zoo via `pip install unsloth_zoo` Also error = {str(e)}")
 pass
 
 from unsloth_zoo.device_type import (

unsloth/models/vision.py

@@ -538,7 +538,14 @@ def from_pretrained(
             if hasattr(auto_config, "quantization_config"):
                 from transformers.quantizers.auto import AUTO_QUANTIZATION_CONFIG_MAPPING
                 quantization_config = auto_config.quantization_config
-                quantizer = AUTO_QUANTIZATION_CONFIG_MAPPING[quantization_config["quant_method"]]
+                quant_method = quantization_config["quant_method"]
+                # Sometimes bitsandbytes_4bit + bitsandbytes_8bit is provided
+                if quant_method == "bitsandbytes" and "bitsandbytes" not in AUTO_QUANTIZATION_CONFIG_MAPPING:
+                    if "bitsandbytes_4bit" not in AUTO_QUANTIZATION_CONFIG_MAPPING:
+                        raise KeyError("Unsloth: AUTO_QUANTIZATION_CONFIG_MAPPING does not have `bitsandbytes_4bit`")
+                    quantizer = AUTO_QUANTIZATION_CONFIG_MAPPING["bitsandbytes_4bit"]
+                else:
+                    quantizer = AUTO_QUANTIZATION_CONFIG_MAPPING[quant_method]
                 quantizer_kwargs = {}
                 # We cannot dequantize since gpt-oss-20b MXFP4 will now be gpt-oss-20b-BF16
                 if load_in_16bit and "dequantize" in inspect.signature(quantizer).parameters: