Skip to content
Merged
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
28 changes: 20 additions & 8 deletions docs/source/installation.mdx
Original file line number Diff line number Diff line change
Expand Up @@ -17,7 +17,9 @@ We provide official support for NVIDIA GPUs, CPUs, Intel XPUs, and Intel Gaudi.
- [CPU](#cpu)
- [Installation via PyPI](#cpu-pip)
- [Compile from Source](#cpu-compile)
- [AMD ROCm (Preview)](#rocm-preview)
- [AMD ROCm (Preview)](#rocm)
- [Installation via PyPI](#rocm-pip)
- [Compile from Source](#rocm-compile)
- [Preview Wheels](#preview-wheels)

## System Requirements[[requirements]]
Expand Down Expand Up @@ -49,9 +51,9 @@ The currently distributed `bitsandbytes` packages are built with the following c
|--------------------|------------------|----------------------|--------------
| **Linux x86-64** | 11.8 - 12.6 | GCC 11.2 | sm60, sm70, sm75, sm80, sm86, sm89, sm90
| **Linux x86-64** | 12.8 - 12.9 | GCC 11.2 | sm70, sm75, sm80, sm86, sm89, sm90, sm100, sm120
| **Linux x86-64** | 13.0 | GCC 11.2 | sm75, sm80, sm86, sm89, sm90, sm100, sm110, sm120
| **Linux x86-64** | 13.0 | GCC 11.2 | sm75, sm80, sm86, sm89, sm90, sm100, sm120
| **Linux aarch64** | 11.8 - 12.6 | GCC 11.2 | sm75, sm80, sm90
| **Linux aarch64** | 12.8 - 13.0 | GCC 11.2 | sm75, sm80, sm90, sm100, sm120
| **Linux aarch64** | 12.8 - 13.0 | GCC 11.2 | sm75, sm80, sm90, sm100, sm110, sm120, sm121
| **Windows x86-64** | 11.8 - 12.6 | MSVC 19.43+ (VS2022) | sm50, sm60, sm75, sm80, sm86, sm89, sm90
| **Windows x86-64** | 12.8 - 12.9 | MSVC 19.43+ (VS2022) | sm70, sm75, sm80, sm86, sm89, sm90, sm100, sm120
| **Windows x86-64** | 13.0 | MSVC 19.43+ (VS2022) | sm75, sm80, sm86, sm89, sm90, sm100, sm120
Expand Down Expand Up @@ -183,12 +185,16 @@ pip install -e .

## AMD ROCm (Preview)[[rocm]]

* Support for AMD GPUs is currently in a preview state.
* All features are supported for consumer RDNA devices.
* The Data Center products currently lack support for the default 4bit blocksize of 64, and as such default to 128.
* A compatible PyTorch version with AMD ROCm support is required. It is recommended to use the latest stable release. See [PyTorch on ROCm](https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/3rd-party/pytorch-install.html) for guidance.
* ROCm support is currently only available in our preview wheels or when building from source.

### Preview Wheels from `main`[[rocm-preview]]
### Installation from PyPI[[rocm-pip]]

The currently distributed preview `bitsandbytes` are built with the following configurations:
This is the most straightforward and recommended installation option.

The currently distributed `bitsandbytes` are built with the following configurations:

| **OS** | **ROCm** | **Targets**
|--------------------|----------|---------------------------------------------------------------------|
Expand All @@ -200,11 +206,17 @@ The currently distributed preview `bitsandbytes` are built with the following co

**Windows is not currently supported.**

Please see [Preview Wheels](#preview-wheels) for installation instructions.
Use `pip` or `uv` to install the latest release:

```bash
pip install bitsandbytes
```

### Compile from Source[[rocm-compile]]

bitsandbytes can be compiled from ROCm 6.1 - ROCm 7.0.
bitsandbytes can be compiled from ROCm 6.2 - ROCm 7.1.

To compile from source, you need CMake >= **3.31.6**.

```bash
# Install bitsandbytes from source
Expand Down