> ## Documentation Index
> Fetch the complete documentation index at: https://mintlify.com/AlexsJones/llmfit/llms.txt
> Use this file to discover all available pages before exploring further.

# Installation

> Install llmfit on macOS, Linux, or Windows

llmfit is distributed as a single binary with no runtime dependencies. Choose your preferred installation method below.

## macOS

### Homebrew (recommended)

Install via Homebrew:

```bash theme={null}
brew install llmfit
```

This installs the latest stable release and keeps llmfit updated with `brew upgrade`.

### curl install script

Alternatively, use the quick install script:

```bash theme={null}
curl -fsSL https://llmfit.axjns.dev/install.sh | sh
```

This downloads the latest release binary from GitHub and installs it to `/usr/local/bin`.

**Install without sudo** (installs to `~/.local/bin`):

```bash theme={null}
curl -fsSL https://llmfit.axjns.dev/install.sh | sh -s -- --local
```

<Note>
  Make sure `~/.local/bin` is in your `$PATH`. Add `export PATH="$HOME/.local/bin:$PATH"` to your shell profile if needed.
</Note>

## Linux

### curl install script (recommended)

Install via the quick install script:

```bash theme={null}
curl -fsSL https://llmfit.axjns.dev/install.sh | sh
```

This downloads the latest release binary and installs it to `/usr/local/bin` (requires sudo).

**Install without sudo** (installs to `~/.local/bin`):

```bash theme={null}
curl -fsSL https://llmfit.axjns.dev/install.sh | sh -s -- --local
```

<Note>
  Make sure `~/.local/bin` is in your `$PATH`. Add `export PATH="$HOME/.local/bin:$PATH"` to your `.bashrc` or `.zshrc`.
</Note>

### Manual installation

1. Download the latest release for your architecture from [GitHub Releases](https://github.com/AlexsJones/llmfit/releases/latest)
2. Extract the tarball:
   ```bash theme={null}
   tar -xzf llmfit-*-unknown-linux-musl.tar.gz
   ```
3. Move the binary to a directory in your `$PATH`:
   ```bash theme={null}
   sudo mv llmfit /usr/local/bin/
   # Or without sudo:
   mkdir -p ~/.local/bin && mv llmfit ~/.local/bin/
   ```
4. Verify installation:
   ```bash theme={null}
   llmfit --version
   ```

## Windows

### Scoop (recommended)

Install via [Scoop](https://scoop.sh/):

```powershell theme={null}
scoop install llmfit
```

If Scoop is not installed, follow the [Scoop installation guide](https://scoop.sh/).

### Manual installation

1. Download the latest Windows release from [GitHub Releases](https://github.com/AlexsJones/llmfit/releases/latest)
2. Extract the ZIP archive
3. Add the extracted directory to your `PATH` environment variable
4. Open a new terminal and verify:
   ```powershell theme={null}
   llmfit --version
   ```

## Build from source

Build the latest development version from source using Rust:

<Steps>
  <Step title="Install Rust">
    If you don't have Rust installed, get it from [rustup.rs](https://rustup.rs/):

    ```bash theme={null}
    curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh
    ```

    Ensure you have Rust 1.85+ (edition 2024 support).
  </Step>

  <Step title="Clone and build">
    Clone the repository and build the release binary:

    ```bash theme={null}
    git clone https://github.com/AlexsJones/llmfit.git
    cd llmfit
    cargo build --release
    ```

    The compiled binary will be at `target/release/llmfit`.
  </Step>

  <Step title="Install the binary">
    Move the binary to a directory in your `$PATH`:

    ```bash theme={null}
    # macOS/Linux
    sudo mv target/release/llmfit /usr/local/bin/

    # Or without sudo
    mkdir -p ~/.local/bin && mv target/release/llmfit ~/.local/bin/
    ```

    On Windows, move `target\release\llmfit.exe` to a directory in your `PATH`.
  </Step>

  <Step title="Verify installation">
    Confirm llmfit is working:

    ```bash theme={null}
    llmfit --version
    ```
  </Step>
</Steps>

## System requirements

### Minimum requirements

* **OS**: Linux (x86\_64, aarch64), macOS (Intel, Apple Silicon), or Windows (x86\_64)
* **RAM**: No specific requirement (llmfit itself uses less than 50 MB)
* **Disk**: \~10 MB for the binary

### Optional: GPU detection tools

For GPU detection to work, install the appropriate tool for your hardware:

| GPU Vendor    | Detection Tool    | Purpose                             |
| ------------- | ----------------- | ----------------------------------- |
| NVIDIA        | `nvidia-smi`      | VRAM reporting, multi-GPU support   |
| AMD           | `rocm-smi`        | ROCm GPU detection                  |
| Intel Arc     | sysfs, `lspci`    | Discrete/integrated GPU detection   |
| Apple Silicon | `system_profiler` | Unified memory reporting (built-in) |
| Ascend        | `npu-smi`         | NPU detection                       |

<Note>
  GPU detection is best-effort. If autodetection fails, use `llmfit --memory=24G` to manually specify your VRAM size.
</Note>

### Optional: Runtime providers

To download and run models from the TUI, install a runtime provider:

* **Ollama**: Download from [ollama.com](https://ollama.com) (macOS, Linux, Windows)
* **llama.cpp**: Build from [github.com/ggml-org/llama.cpp](https://github.com/ggml-org/llama.cpp) or install via package manager
* **MLX**: Install via pip on Apple Silicon: `pip install mlx-lm`

llmfit works without these tools installed — you just won't be able to download or detect installed models.

## Verify installation

After installing, verify llmfit is working:

```bash theme={null}
# Check version
llmfit --version

# Display system specs
llmfit system

# Launch TUI
llmfit
```

You should see your hardware specs (RAM, CPU cores, GPU name and VRAM).

## Troubleshooting

### Command not found

If you see `command not found: llmfit`, the binary is not in your `$PATH`:

1. Check where llmfit is installed:
   ```bash theme={null}
   which llmfit
   ```
2. If it's in `~/.local/bin`, add this to your shell profile (`~/.bashrc`, `~/.zshrc`, etc.):
   ```bash theme={null}
   export PATH="$HOME/.local/bin:$PATH"
   ```
3. Reload your shell:
   ```bash theme={null}
   source ~/.bashrc  # or ~/.zshrc
   ```

### GPU not detected

If your GPU is not detected:

1. Check if the detection tool is installed:
   ```bash theme={null}
   # NVIDIA
   nvidia-smi

   # AMD
   rocm-smi
   ```
2. If the tool is missing, install it via your package manager or GPU vendor's website
3. If detection still fails, manually specify VRAM:
   ```bash theme={null}
   llmfit --memory=24G
   ```

### Incorrect VRAM reported

On some systems (VMs, passthrough setups, broken drivers), VRAM autodetection may fail or report incorrect values. Override with `--memory`:

```bash theme={null}
# Specify 32 GB VRAM
llmfit --memory=32G

# Works with all modes
llmfit --memory=24G --cli
llmfit --memory=24G fit --perfect -n 5
```

Accepted suffixes: `G`/`GB`/`GiB` (gigabytes), `M`/`MB`/`MiB` (megabytes), `T`/`TB`/`TiB` (terabytes). Case-insensitive.

### Permission denied (install script)

If the install script fails with permission denied:

1. Try the `--local` flag to install without sudo:
   ```bash theme={null}
   curl -fsSL https://llmfit.axjns.dev/install.sh | sh -s -- --local
   ```
2. Make sure `~/.local/bin` is in your `$PATH`

## Update llmfit

Update to the latest version using the same method you used to install:

<Tabs>
  <Tab title="Homebrew (macOS/Linux)">
    ```bash theme={null}
    brew upgrade llmfit
    ```
  </Tab>

  <Tab title="Scoop (Windows)">
    ```powershell theme={null}
    scoop update llmfit
    ```
  </Tab>

  <Tab title="curl install script">
    ```bash theme={null}
    curl -fsSL https://llmfit.axjns.dev/install.sh | sh
    ```

    The script always installs the latest release.
  </Tab>

  <Tab title="From source">
    ```bash theme={null}
    cd llmfit
    git pull
    cargo build --release
    sudo mv target/release/llmfit /usr/local/bin/
    ```
  </Tab>
</Tabs>

## Uninstall

Remove llmfit by deleting the binary:

<Tabs>
  <Tab title="Homebrew">
    ```bash theme={null}
    brew uninstall llmfit
    ```
  </Tab>

  <Tab title="Scoop">
    ```powershell theme={null}
    scoop uninstall llmfit
    ```
  </Tab>

  <Tab title="Manual">
    ```bash theme={null}
    # Find where it's installed
    which llmfit

    # Remove the binary
    sudo rm /usr/local/bin/llmfit
    # Or if installed locally:
    rm ~/.local/bin/llmfit
    ```
  </Tab>
</Tabs>

Optionally, remove config files:

```bash theme={null}
rm -rf ~/.config/llmfit
```

## Next steps

<CardGroup cols={2}>
  <Card title="Quickstart" icon="rocket" href="/quickstart">
    Get from installation to first successful run
  </Card>

  <Card title="TUI Mode" icon="terminal" href="/guides/tui-mode">
    Learn keyboard shortcuts and navigation
  </Card>

  <Card title="CLI Mode" icon="code" href="/guides/cli-mode">
    Use llmfit in scripts and automation
  </Card>

  <Card title="System Commands" icon="microchip" href="/api/commands/system">
    Check detected hardware specs
  </Card>
</CardGroup>