> For the complete documentation index, see [llms.txt](https://docs.modular.com/llms.txt). > Markdown versions of all pages are available by appending .md to any URL (e.g. /max/get-started.md). # Packages All the Modular APIs and tools (including MAX and Mojo) are included in a single Python/Conda package named `modular`. This page explains what's in the package and shows how to install it. If you just want to get started with Modular, instead see our [quickstart guide](https://docs.modular.com/max/get-started.md). ## Package managers You can install `modular` as a Python or Conda package. The install instructions below include commands for popular package managers such as `pixi`, `uv`, `pip`, and `conda`, but you can also use something else. Traditional tools like `pip` and `conda` might be necessary for existing production environments or Docker containers, but for your local development, we recommend using [Pixi](https://pixi.sh/latest/). Pixi is both a package manager and virtual environment manager, which alone makes development a lot easier, but it's also language agnostic, extremely fast, and includes lock files to easily reproduce your project environment. You'll notice that our [GitHub code examples](https://github.com/modular/modular/tree/main/max/examples) include a `pixi.toml` file. This file configures the environment to make sure we all use the same packages and get the same results—you just need to install `pixi`. So if you're not set on using a particular package manager, we suggest you try `pixi`. If you haven't used it before, check out our [Pixi basics guide](https://docs.modular.com/pixi.md) or the [official Pixi docs](https://pixi.sh/latest/). ## Install To get the latest improvements and new features, we recommend installing our nightly build, which we release several times a week. If you want a better tested but older version, you can install a stable build. (Each release is described in the [changelog](https://docs.modular.com/max/changelog.md).) System requirements: [Read the requirements](https://docs.modular.com/max/packages.md#system-requirements) :::tip We recommend installing with `pixi` for the most reliable experience. ::: The `modular` package installs MAX, Mojo, and other package dependencies. :::note GitHub stable branch If you're using a stable build and want to clone the [Modular repo](https://github.com/modular/modular), make sure you checkout the `stable` branch (because the `main` branch includes the latest nightly code). For example: ```sh git clone -b stable https://github.com/modular/modular.git ``` ::: ## Uninstall **pixi:** You can uninstall `modular` from your virtual environment with this command: ```sh pixi remove modular ``` To deactivate your virtual environment, run: ```sh exit ``` --- **uv:** You can uninstall `modular` from your virtual environment with the following command: ```sh uv pip uninstall modular ``` To deactivate your virtual environment, run: ```sh deactivate ``` --- **pip:** You can uninstall `modular` from your virtual environment with the following command: ```sh pip uninstall modular ``` To deactivate your virtual environment, run: ```sh deactivate ``` --- **conda:** You can uninstall `modular` from your virtual environment with this command: ```sh conda remove modular ``` To deactivate your virtual environment, run: ```sh deactivate ``` ## What's included Here's a summary of what's in the `modular` package. **conda/pixi:** The `modular` Conda package installs the following: - MAX tools and libraries: - [`max` CLI](https://docs.modular.com/max/cli.md) - [`max` Python library](https://docs.modular.com/max/api/python.md) - [`max` Mojo library](https://docs.modular.com/max/api/kernels/comm.md) - [MAX Engine C API](https://docs.modular.com/max/api/c.md) - The `mojo` package: - [`mojo` CLI](https://mojolang.org/docs/cli/) (includes the Mojo compiler) - [Mojo standard library](https://mojolang.org/docs/std/) - Mojo language server (LSP) for IDE/editor integration - [Mojo debugger](https://mojolang.org/docs/tools/debugging/) (includes LLDB) - [Mojo code formatter](https://mojolang.org/docs/cli/format/) - [Mojo REPL](https://mojolang.org/docs/cli/repl/) `pixi` known issues: - You might encounter issues if you invoke `pixi` within a `conda` virtual environment. It's best if you don't mix the two tools. --- **pip/uv:** The `modular` Python wheel installs the following: - MAX tools and libraries: - [`max` CLI](https://docs.modular.com/max/cli.md) - [`max` Python library](https://docs.modular.com/max/api/python.md) - [`max` Mojo library](https://docs.modular.com/max/api/kernels/comm.md) - The `mojo` package: - [`mojo` CLI](https://mojolang.org/docs/cli/) (includes the Mojo compiler) - [Mojo standard library](https://mojolang.org/docs/std/) - Mojo language server (LSP) for IDE/editor integration - [Mojo debugger](https://mojolang.org/docs/tools/debugging/) (includes LLDB) - [Mojo code formatter](https://mojolang.org/docs/cli/format/) - [Mojo REPL](https://mojolang.org/docs/cli/repl/) :::note Note For information about the `mojo` package, see the [Mojo install guide](https://mojolang.org/docs/manual/install/). ::: ## System requirements To install `modular`, your system must meet these specifications. **Mac:** - macOS Sequoia (15) or later. - Apple silicon (M1–M5 processor). - Python 3.10–3.14. - Xcode or Xcode Command Line Tools 16 or later. - Apple silicon GPU support is available for Mojo GPU programming. See [GPU compatibility](#gpu-compatibility) for details. You may need to run `xcodebuild -downloadComponent MetalToolchain` to install the Metal utilities required for GPU programming. --- **Linux:** - glibc 2.34 or later (for example, Ubuntu 22.04 LTS or later). :::note Officially supported distribution Modular tests MAX on Ubuntu 22.04 LTS or later. Other Linux distributions that meet the glibc 2.34+ requirement are expected to work but aren't continuously tested. ::: - x86-64-v3 (Haswell-class or newer; CPUs from approximately 2013 onward) or ARM64 Neoverse N1 or newer (for example, AWS Graviton2 and later). :::note x86-64-v3 CPU instruction sets The [x86-64-v3 microarchitecture level](https://github.com/llvm/llvm-project/blob/main/clang/docs/UsersManual.rst#x86) requires AVX, AVX2, BMI1, BMI2, F16C, FMA, LZCNT, MOVBE, and XSAVE instructions. To verify your x86-64 CPU on Linux, run `cat /proc/cpuinfo | grep flags` and confirm those flags are present. This requirement doesn't apply to ARM64 hosts. ::: - 8 GiB RAM minimum for Mojo development. MAX inference and model serving require significantly more memory, with the exact amount varying by model. For the catalog of supported models, see the [supported models](https://docs.modular.com/max/models.md) page; for a specific model, check its Hugging Face page for memory details. - Python 3.10–3.14. - A C compiler (such as `cc`, `gcc`, or `clang`)—used as a linker. - To use GPUs, see the [GPU compatibility](#gpu-compatibility). --- **Windows:** Windows is not officially supported at this time. In the meantime, you can try MAX on Windows [with WSL](https://learn.microsoft.com/en-us/windows/wsl/install), using a compatible version of Ubuntu (see our requirements for Linux). ### GPU compatibility MAX supports both CPUs and GPUs, so you don't always need a GPU to serve a model. But if you want to accelerate inference with GPUs or program for GPUs with Mojo, the following GPUs are compatible. #### NVIDIA GPUs
GPU Architecture
Tested for serving
B200 Blackwell
Known compatible for development
B300 Blackwell
B100 Blackwell
DGX Spark Blackwell
H200 Hopper
H100 Hopper
L4 Ada Lovelace
L40 Ada Lovelace
A100 Ampere
A10 Ampere
A1000 Ampere
RTX 50XX series Blackwell
RTX 40XX series Ada Lovelace
RTX 30XX series Ampere
Jetson Orin / Orin Nano Ampere
Jetson Thor Blackwell
T4 Turing
RTX 20XX series Turing
Software requirement: NVIDIA GPU driver 580 or later. Check your version with [`nvidia-smi`](https://developer.nvidia.com/system-management-interface). To update, see the [NVIDIA driver docs](https://www.nvidia.com/en-us/drivers/). If your driver is older than 580, set the `MODULAR_NVPTX_COMPILER_PATH` environment variable to a system `ptxas` binary (for example, `export MODULAR_NVPTX_COMPILER_PATH=/usr/local/cuda/bin/ptxas`). #### AMD GPUs
GPU Architecture
Tested for serving
MI355X CDNA4
MI300X CDNA3
MI325X CDNA3
Known compatible for development
MI250X CDNA2
Radeon RX 9070 RDNA4
Radeon RX 9060 RDNA4
Radeon 880M / 890M RDNA3.5
Radeon 860M RDNA3.5
Radeon 8060S RDNA3.5
Radeon RX 7900 RDNA3
Radeon RX 7800 / 7700 RDNA3
Radeon RX 7600 RDNA3
Radeon 780M RDNA3
Radeon RX 6900 RDNA2
Software requirement: AMD GPU driver 6.3.3 or later (MI355X requires ROCm 7.0 or later). For installation, see the [Ubuntu native install guide](https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/install-methods/package-manager/package-manager-ubuntu.html). #### Apple silicon GPUs Apple silicon GPU support (M1–M5) is functional for GPU programming with Mojo. Large GenAI model inference via MAX isn't yet available on Apple silicon. For details, see the [Mojo system requirements](https://mojolang.org/docs/requirements/#apple-silicon-gpus). :::note Notes - Many GPUs are available in variants with different amounts of memory, and each AI model has different memory requirements. Even if your GPU architecture is listed as compatible, confirm that the available memory is sufficient for the model you're using. - MAX can serve models on either CPU or GPU, but some models require one or more GPUs. When you browse our [supported models](https://docs.modular.com/max/models.md), you can check for multi-GPU support. - For comprehensive GPU hardware tables, arch targets, and troubleshooting, see the [Mojo system requirements](https://mojolang.org/docs/requirements/#gpu-compatibility) page. - To deploy MAX in containers, see the [deployment guide](https://docs.modular.com/max/deploy/cloud.md). :::