crackedai-connect 0.2.0

Local runtime for crackedaicode.com — run ML code on your own machine

crackedai-connect

Local runtime for crackedaicode.com — execute ML code on your own machine using your own GPU.

Install & Run

pip install crackedai-connect
crackedai-connect

That's it. On first run, the CLI auto-detects your hardware and installs the right versions of PyTorch and JAX:

• NVIDIA GPU — installs PyTorch + JAX with CUDA support (detects your CUDA version)
• Apple Silicon — installs PyTorch with MPS + JAX
• CPU only — installs lightweight CPU-only versions

$ crackedai-connect

  CrackedAI Runtime v0.1.0

  Missing: PyTorch, JAX
  Detecting hardware...
  Found NVIDIA GPU (CUDA 12.4)
  Installing PyTorch (CUDA 12)... done
  Installing JAX (CUDA 12)... done

  Detected: NVIDIA RTX 4090 | Linux
    pytorch 2.3.0 ✓ (CUDA accelerated)
    jax 0.4.30 ✓ (cuda)
    numpy 1.26.4 ✓

  Listening on http://127.0.0.1:52341
  Open crackedaicode.com to start solving problems.

Then open crackedaicode.com — the website automatically detects the running runtime.

What This Does

When you click "Run" or "Submit" on crackedaicode.com, your code executes locally on your machine — not on a remote server:

• Your own GPU — CUDA, Apple Silicon MPS, or CPU
• Your own packages — whatever versions the installer sets up
• Zero latency — no round-trip to a server
• Full privacy — code never leaves your machine

How It Works

Browser (crackedaicode.com)  <-->  localhost:52341 (crackedai-connect)
                                          |
                                 Python subprocess
                                    (your code)

Already Have PyTorch/JAX?

If you already have them installed, crackedai-connect will detect and use your existing installation. It only installs frameworks if they're missing.

For Contributors

git clone https://github.com/anupa/cracked-ai-connect.git
cd cracked-ai-connect
pixi install
pixi run test
pixi run connect

License

MIT