Nightly (#3169)

* Fix mamba

* Update loader.py

* Update vision.py

* Update loader.py

* Filter vLLM standby logs (#3131)

* filter vLLM standby logs

* safeguard standby logger patch

* Update unsloth/models/_utils.py

* Update unsloth/models/_utils.py

* Update unsloth/models/_utils.py

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update loader.py

* Add scaler

* Update llama.py

* Update _utils.py

* Versioning

* GPT OSS fix

* GPT OSS fix

* Update loader.py

* Update vision.py

* Update vision.py

* Update loader.py

* Update vision.py

* Update vision.py

* Update llama.py

* Update llama.py

* Update llama.py

* Versioning

* Update mapper.py

---------

Co-authored-by: Datta Nimmaturi <venkatadattasainimmaturi@gmail.com>

pyproject.toml

@@ -37,7 +37,7 @@ triton = [
 ]
 
 huggingface = [
-    "unsloth_zoo>=2025.8.4",
+    "unsloth_zoo>=2025.8.5",
     "packaging",
     "tyro",
     "transformers>=4.51.3,!=4.47.0,!=4.52.0,!=4.52.1,!=4.52.2,!=4.52.3,!=4.53.0",
@@ -384,7 +384,7 @@ colab-ampere-torch220 = [
     "flash-attn>=2.6.3",
 ]
 colab-new = [
-    "unsloth_zoo>=2025.8.4",
+    "unsloth_zoo>=2025.8.5",
     "packaging",
     "tyro",
     "transformers>=4.51.3,!=4.47.0,!=4.52.0,!=4.52.1,!=4.52.2,!=4.52.3,!=4.53.0",

unsloth/models/_utils.py

@@ -12,7 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-__version__ = "2025.8.5"
+__version__ = "2025.8.6"
 
 __all__ = [
     "SUPPORTS_BFLOAT16",

unsloth/models/llama.py

@@ -701,8 +701,9 @@ def LlamaModel_fast_forward(
     # Fix out of bounds tokenization
     if hasattr(self, "max_seq_length"):
         if seq_length > self.max_seq_length:
+            shape = input_ids.shape if input_ids is not None else inputs_embeds.shape
             logger.warning_once(
-                f"Unsloth: Input IDs of length {seq_length} > the model's max sequence length of {self.max_seq_length}.\n"\
+                f"Unsloth: Input IDs of shape {shape} with length {seq_length} > the model's max sequence length of {self.max_seq_length}.\n"\
                 "We shall truncate it ourselves. It's imperative if you correct this issue first."
             )
         if input_ids is not None:

unsloth/models/mapper.py

@@ -941,6 +941,16 @@ __INT_TO_FLOAT_MAPPER = \
         "Qwen/Qwen3-4B-Thinking-2507",
         "unsloth/Qwen3-4B-Thinking-2507-bnb-4bit",
     ),
+    "unsloth/gemma-3-270m-it-unsloth-bnb-4bit" : (
+        "unsloth/gemma-3-270m-it",
+        "google/gemma-3-270m-it",
+        "unsloth/gemma-3-270m-it-bnb-4bit",
+    ),
+    "unsloth/gemma-3-270m-unsloth-bnb-4bit" : (
+        "unsloth/gemma-3-270m",
+        "google/gemma-3-270m",
+        "unsloth/gemma-3-270m-bnb-4bit",
+    ),
 }
 
 INT_TO_FLOAT_MAPPER  = {}