You are not logged in.
Website: https://usehax.dev/
Source code: https://github.com/OleksandrChekhovskyi/hax (MIT-licensed)
AUR package: hax
Today, most of the coding agent implementations use bloated JS or Python stacks, consume a lot of system resources (typically hundreds of MBs), don't respect XDG paths, take over the terminal (messing up the scrollback), etc.
I wanted to make something different, something that would feel like a proper Unix tool. hax is the result of that work.
Consider this as a lean alternative to Claude Code, Codex, OpenCode, pi, and others.
Key features
- Lightweight by design — A single native C binary with a small dependency set. Starts instantly and uses only a few MB of memory, leaving more RAM for local models.
- Local models are first-class — Start `llama-server -m [model].gguf`, then `hax --provider llama.cpp`, and hax auto-discovers the model and runtime capabilities. No custom provider config block is needed for the default setup.
- Respects your terminal — Streaming Markdown and live tool output, reflowed for display in the terminal. Only redraws the current streaming line or the input area, native scrollback is preserved. Does not take over or mess with your terminal.
- Inspectable — See exactly what was sent to the model and what it replied in a usable transcript view (Ctrl+T). Optionally collect a detailed wire protocol trace.
- Broad provider support — OpenAI (+compatible), Anthropic (+compatible), Codex (via a ChatGPT subscription), OpenRouter, OpenCode Zen/Go, llama.cpp, and custom endpoints.
- Well-behaved Unix tool — XDG paths, clean stdout in `-p` one-shot mode with resume hints on stderr, plain-text config and session files, composition via subprocesses instead of plugins.
Offline