# MLX

## Docs

- [Array](https://mintlify.wiki/ml-explore/mlx/api/array.md): Core array class and methods for MLX arrays
- [CUDA Backend](https://mintlify.wiki/ml-explore/mlx/api/cuda.md): Check CUDA availability and use MLX on NVIDIA GPUs
- [Data Types](https://mintlify.wiki/ml-explore/mlx/api/data-types.md): Supported data types and type inspection utilities in MLX
- [Devices and Streams](https://mintlify.wiki/ml-explore/mlx/api/devices-streams.md): Managing computation devices and execution streams in MLX
- [Distributed Communication](https://mintlify.wiki/ml-explore/mlx/api/distributed.md): Multi-node and multi-GPU communication primitives for distributed training and inference
- [Export Functions](https://mintlify.wiki/ml-explore/mlx/api/export.md): Export and import MLX computation graphs for deployment and optimization
- [Fast Operations](https://mintlify.wiki/ml-explore/mlx/api/fast.md): High-performance optimized operations for common deep learning patterns
- [Fast Fourier Transform](https://mintlify.wiki/ml-explore/mlx/api/fft.md): Discrete Fourier Transform operations in MLX
- [Linear Algebra](https://mintlify.wiki/ml-explore/mlx/api/linalg.md): Linear algebra operations in MLX
- [Memory Management](https://mintlify.wiki/ml-explore/mlx/api/memory.md): Monitor and control GPU memory usage in MLX
- [Metal Backend](https://mintlify.wiki/ml-explore/mlx/api/metal.md): Access Metal-specific functionality and GPU information on Apple devices
- [Neural Networks](https://mintlify.wiki/ml-explore/mlx/api/nn.md): Build neural networks with MLX using the nn module
- [Operations](https://mintlify.wiki/ml-explore/mlx/api/ops.md): Core array operations and mathematical functions in MLX
- [Optimizers](https://mintlify.wiki/ml-explore/mlx/api/optimizers.md): Optimize model parameters with MLX optimizers
- [Random Sampling](https://mintlify.wiki/ml-explore/mlx/api/random.md): Random sampling functions for generating pseudo-random numbers in MLX
- [Function Transforms](https://mintlify.wiki/ml-explore/mlx/api/transforms.md): Automatic differentiation and function transformation operations in MLX
- [Tree Utils](https://mintlify.wiki/ml-explore/mlx/api/tree-utils.md): Utilities for working with nested Python data structures
- [Compilation](https://mintlify.wiki/ml-explore/mlx/concepts/compilation.md): Learn how to use MLX's compile function to optimize computation graphs for better performance
- [Function Transforms](https://mintlify.wiki/ml-explore/mlx/concepts/function-transforms.md): Learn about composable function transformations for automatic differentiation, vectorization, and optimization
- [Lazy Evaluation](https://mintlify.wiki/ml-explore/mlx/concepts/lazy-evaluation.md): Understand how MLX uses lazy evaluation to build compute graphs and optimize performance
- [Unified Memory](https://mintlify.wiki/ml-explore/mlx/concepts/unified-memory.md): Learn how MLX takes advantage of Apple Silicon's unified memory architecture
- [Building C++ Extensions](https://mintlify.wiki/ml-explore/mlx/cpp/extensions.md): Create custom MLX operations with C++ primitives and Metal kernels
- [Custom Metal Kernels](https://mintlify.wiki/ml-explore/mlx/cpp/metal-kernels.md): Write high-performance GPU kernels for MLX using Metal
- [C++ Operations Reference](https://mintlify.wiki/ml-explore/mlx/cpp/ops.md): Complete reference for MLX C++ array operations and functions
- [C++ API Overview](https://mintlify.wiki/ml-explore/mlx/cpp/overview.md): Introduction to using MLX with C++ for high-performance machine learning
- [Using MLX in C++](https://mintlify.wiki/ml-explore/mlx/cpp/usage.md): Guide to integrating MLX into C++ projects with CMake
- [Data Parallelism](https://mintlify.wiki/ml-explore/mlx/examples/data-parallelism.md): Implement distributed data parallel training across multiple devices using MLX distributed primitives
- [Linear Regression](https://mintlify.wiki/ml-explore/mlx/examples/linear-regression.md): Build a basic linear regression model using MLX with automatic differentiation and stochastic gradient descent
- [LLM Inference](https://mintlify.wiki/ml-explore/mlx/examples/llama-inference.md): Implement efficient inference for Llama-family transformer models using MLX neural network modules
- [Multi-Layer Perceptron](https://mintlify.wiki/ml-explore/mlx/examples/neural-network.md): Build and train a neural network for MNIST classification using mlx.nn modules and optimizers
- [Tensor Parallelism](https://mintlify.wiki/ml-explore/mlx/examples/tensor-parallelism.md): Shard model parameters across multiple devices using MLX tensor parallelism for large language model inference
- [Distributed Communication](https://mintlify.wiki/ml-explore/mlx/guides/distributed.md): Learn how to use MLX's distributed communication features for multi-machine training and inference across MPI, Ring, JACCL, and NCCL backends
- [Exporting Functions](https://mintlify.wiki/ml-explore/mlx/guides/export.md): Learn how to export and import MLX functions to run computations across different front-ends like Python and C++
- [Indexing Arrays](https://mintlify.wiki/ml-explore/mlx/guides/indexing.md): Learn how to index and slice MLX arrays, including advanced indexing techniques and differences from NumPy
- [Conversion to NumPy and Other Frameworks](https://mintlify.wiki/ml-explore/mlx/guides/numpy-comparison.md): Learn how to convert MLX arrays to and from NumPy, PyTorch, JAX, and TensorFlow using the Python Buffer Protocol and DLPack
- [Saving and Loading Arrays](https://mintlify.wiki/ml-explore/mlx/guides/saving-loading.md): Learn how to serialize and deserialize MLX arrays using NumPy, Safetensors, and GGUF formats
- [Using Streams](https://mintlify.wiki/ml-explore/mlx/guides/streams.md): Learn how to specify and use streams for controlling operation execution in MLX
- [Installation](https://mintlify.wiki/ml-explore/mlx/installation.md): Install MLX on macOS with Apple silicon or Linux with CUDA/CPU support
- [Introduction to MLX](https://mintlify.wiki/ml-explore/mlx/introduction.md): An array framework for machine learning on Apple silicon, designed by machine learning researchers for machine learning researchers
- [Quickstart](https://mintlify.wiki/ml-explore/mlx/quickstart.md): Get up and running with MLX in minutes