Add mlx support to BatchEncoding.convert_to_tensors (#29406)

* Add mlx support

* Fix import order and use def instead of lambda

* Another fix for ruff format :)

* Add detecting mlx from repr, add is_mlx_array

src/transformers/tokenization_utils_base.py

@@ -48,6 +48,7 @@
     extract_commit_hash,
     is_flax_available,
     is_jax_tensor,
+    is_mlx_available,
     is_numpy_array,
     is_offline_mode,
     is_remote_url,
@@ -726,6 +727,16 @@ def as_tensor(value, dtype=None):
 
             as_tensor = jnp.array
             is_tensor = is_jax_tensor
+
+        elif tensor_type == TensorType.MLX:
+            if not is_mlx_available():
+                raise ImportError("Unable to convert output to MLX tensors format, MLX is not installed.")
+            import mlx.core as mx
+
+            as_tensor = mx.array
+
+            def is_tensor(obj):
+                return isinstance(obj, mx.array)
         else:
 
             def as_tensor(value, dtype=None):

src/transformers/utils/__init__.py

@@ -134,6 +134,7 @@
     is_keras_nlp_available,
     is_levenshtein_available,
     is_librosa_available,
+    is_mlx_available,
     is_natten_available,
     is_ninja_available,
     is_nltk_available,

src/transformers/utils/generic.py

@@ -28,7 +28,14 @@
 import numpy as np
 from packaging import version
 
-from .import_utils import get_torch_version, is_flax_available, is_tf_available, is_torch_available, is_torch_fx_proxy
+from .import_utils import (
+    get_torch_version,
+    is_flax_available,
+    is_mlx_available,
+    is_tf_available,
+    is_torch_available,
+    is_torch_fx_proxy,
+)
 
 
 if is_flax_available():
@@ -87,6 +94,8 @@ def infer_framework_from_repr(x):
         return "jax"
     elif representation.startswith("<class 'numpy."):
         return "np"
+    elif representation.startswith("<class 'mlx."):
+        return "mlx"
 
 
 def _get_frameworks_and_test_func(x):
@@ -99,6 +108,7 @@ def _get_frameworks_and_test_func(x):
         "tf": is_tf_tensor,
         "jax": is_jax_tensor,
         "np": is_numpy_array,
+        "mlx": is_mlx_array,
     }
     preferred_framework = infer_framework_from_repr(x)
     # We will test this one first, then numpy, then the others.
@@ -111,8 +121,8 @@ def _get_frameworks_and_test_func(x):
 
 def is_tensor(x):
     """
-    Tests if `x` is a `torch.Tensor`, `tf.Tensor`, `jaxlib.xla_extension.DeviceArray` or `np.ndarray` in the order
-    defined by `infer_framework_from_repr`
+    Tests if `x` is a `torch.Tensor`, `tf.Tensor`, `jaxlib.xla_extension.DeviceArray`, `np.ndarray` or `mlx.array`
+    in the order defined by `infer_framework_from_repr`
     """
     # This gives us a smart order to test the frameworks with the corresponding tests.
     framework_to_test_func = _get_frameworks_and_test_func(x)
@@ -231,6 +241,19 @@ def is_jax_tensor(x):
     return False if not is_flax_available() else _is_jax(x)
 
 
+def _is_mlx(x):
+    import mx.core as mx
+
+    return isinstance(x, mx.array)
+
+
+def is_mlx_array(x):
+    """
+    Tests if `x` is a mlx array or not. Safe to call even when mlx is not installed.
+    """
+    return False if not is_mlx_available() else _is_mlx(x)
+
+
 def to_py_obj(obj):
     """
     Convert a TensorFlow tensor, PyTorch tensor, Numpy array or python list to a python list.
@@ -499,6 +522,7 @@ class TensorType(ExplicitEnum):
     TENSORFLOW = "tf"
     NUMPY = "np"
     JAX = "jax"
+    MLX = "mlx"
 
 
 class ContextManagers:

src/transformers/utils/import_utils.py

@@ -145,6 +145,7 @@ def _is_package_available(pkg_name: str, return_version: bool = False) -> Union[
 _torchaudio_available = _is_package_available("torchaudio")
 _torchdistx_available = _is_package_available("torchdistx")
 _torchvision_available = _is_package_available("torchvision")
+_mlx_available = _is_package_available("mlx")
 
 
 _torch_version = "N/A"
@@ -923,6 +924,10 @@ def is_jinja_available():
     return _jinja_available
 
 
+def is_mlx_available():
+    return _mlx_available
+
+
 # docstyle-ignore
 CV2_IMPORT_ERROR = """
 {0} requires the OpenCV library but it was not found in your environment. You can install it with: