mlex 0.1.2

Safe, idiomatic Rust runtime for Apple MLX: quantized LLM inference (Qwen, Gemma4, NemotronH, DharaAR, ...) with multi-modal support
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30

.. _nn_distributed:

Distributed
-----------

Helper Routines
^^^^^^^^^^^^^^^

The :code:`mlx.nn.layers.distributed` package contains helpful routines to 
create sharded layers from existing :class:`Modules <mlx.nn.Module>`.

.. currentmodule:: mlx.nn.layers.distributed
.. autosummary::
   :toctree: _autosummary

   shard_linear
   shard_inplace

Layers
^^^^^^

.. currentmodule:: mlx.nn
.. autosummary::
   :toctree: _autosummary
   :template: nn-module-template.rst

   AllToShardedLinear
   ShardedToAllLinear
   QuantizedAllToShardedLinear
   QuantizedShardedToAllLinear