Pre-compiled binaries for all major platforms
Released: November 9, 2025 - Tensor Element Access (HITO 8)
Added tensor_get() and tensor_set() for fine-grained tensor element access, enabling advanced architectures like Mixture of Experts (MoE) and dynamic routing mechanisms.
View Release Notescurl -sSf https://charlbase.org/install.sh | shcurl -sSf https://charlbase.org/install.sh | shirm https://charlbase.org/install.ps1 | iexThe installer will automatically detect your platform and download the appropriate binary.
For Ubuntu, Debian, Fedora, Arch Linux, and other x86_64 distributions
| File: | charl-linux-x86_64.tar.gz |
| Version: | v0.4.2 |
| Architecture: | x86_64 (64-bit) |
wget https://github.com/charlcoding-stack/charlcode/releases/download/v0.4.2/charl-linux-x86_64.tar.gz
tar -xzf charl-linux-x86_64.tar.gz
sudo mv charl /usr/local/bin/
charl --versionFor Raspberry Pi, ARM servers, and other aarch64 systems
| File: | charl-linux-arm64.tar.gz |
| Version: | v0.4.2 |
| Architecture: | aarch64 (ARM 64-bit) |
wget https://github.com/charlcoding-stack/charlcode/releases/download/v0.4.2/charl-linux-arm64.tar.gz
tar -xzf charl-linux-arm64.tar.gz
sudo mv charl /usr/local/bin/
charl --versionFor Intel-based Macs (macOS 10.15 Catalina or later)
| File: | charl-macos-x86_64.tar.gz |
| Version: | v0.4.2 |
| Architecture: | x86_64 (Intel 64-bit) |
curl -LO https://github.com/charlcoding-stack/charlcode/releases/download/v0.4.2/charl-macos-x86_64.tar.gz
tar -xzf charl-macos-x86_64.tar.gz
sudo mv charl /usr/local/bin/
charl --versionFor M1, M2, M3, M4 Macs (macOS 11.0 Big Sur or later)
| File: | charl-macos-arm64.tar.gz |
| Version: | v0.4.2 |
| Architecture: | aarch64 (Apple Silicon) |
curl -LO https://github.com/charlcoding-stack/charlcode/releases/download/v0.4.2/charl-macos-arm64.tar.gz
tar -xzf charl-macos-arm64.tar.gz
sudo mv charl /usr/local/bin/
charl --versionFor Windows 10/11 (64-bit)
| File: | charl-windows-x86_64.zip |
| Version: | v0.4.2 |
| Architecture: | x86_64 (64-bit) |
# Download and install using PowerShell
irm https://charlbase.org/install.ps1 | iex
# Or download manually and add to PATH
# Then run: charl.exe --versionAlready have Charl installed? Here's how to update to the latest version (v0.4.2):
Simply re-run the installer:
irm https://charlbase.org/install.ps1 | iexThe installer will automatically detect and replace your existing installation with v0.4.2.
Simply re-run the installer:
curl -sSf https://charlbase.org/install.sh | shThe installer will automatically update your existing installation to v0.4.2.
Prefer to do it manually?
charl --version to verifyRun charl --version to see which version you currently have installed.
For developers who want to compile Charl from source or contribute to development.
# Clone repository
git clone https://github.com/charlcoding-stack/charlcode.git
cd charlcode
# Build release binary
cargo build --release
# Binary location
./target/release/charl --version
# Install system-wide
cargo install --path .# Run all tests (564 tests)
cargo test
# Run with verbose output
cargo test -- --nocapture| Version | Release Date | Highlights | Links |
|---|---|---|---|
| v0.4.2 | Nov 9, 2025 | Tensor element access with tensor_get() and tensor_set() for MoE and dynamic routing (HITO 8) | Release Notes |
| v0.3.0 | Nov 7, 2025 | Full neural network training with autograd, gradient tracking, backpropagation, and optimizers | Release Notes |
| v0.2.0 | Jan 6, 2025 | Complete deep learning framework - 7 neural layers, MLPs, CNNs, Transformers support | Release Notes |
| v0.1.6 | Nov 6, 2025 | Self-update command 'charl update', bug fixes | Release Notes |
| v0.1.3 | Nov 6, 2025 | Multi-platform binaries, automated installers | Release Notes |
| v0.1.0 | Nov 5, 2025 | Initial public release, 34 ML functions, full CLI, REPL, 564 tests passing | Release Notes |