Installation
The fastest way is the Download page, which detects your OS. Or use a one-line installer:
macOS and Linux
curl --proto '=https' --tlsv1.2 -LsSf https://github.com/veermshah/veer-site/releases/latest/download/veer-cli-installer.sh | shWindows (PowerShell)
powershell -ExecutionPolicy Bypass -c "irm https://github.com/veermshah/veer-site/releases/latest/download/veer-cli-installer.ps1 | iex"Windows users can also download the .msi installer from the Download page.
Check it works
Section titled “Check it works”veer --versionveer explain V0101Editor support
Section titled “Editor support”Veer files use the .veer extension. There is no language server yet. Rust or TypeScript syntax highlighting is a reasonable stand-in until there is (see the roadmap).
Running against a real model
Section titled “Running against a real model”veer run uses the built-in simulated engine by default, so you can learn the language with no GPU. To use a real model, start any OpenAI-compatible server and pass --engine http. See Engines. For example, on a Mac with llama.cpp:
llama-server -m qwen2.5-0.5b-instruct-q8_0.gguf --port 8080veer run hello.veer --engine http --endpoint http://localhost:8080 --dialect llamacpp