Skip to content

docs: clarify deterministic quick start requirements #7

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
rustfuture merged 1 commit into from
Oct 2, 2026
Merged

docs: clarify deterministic quick start requirements #7

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 4 additions & 2 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,14 +18,16 @@ RSI (recursive self-improvement) is the research question behind the name. This

## Quick start

The deterministic run uses only the Python standard library.
You need Git and Python 3.10+ ([Python downloads](https://www.python.org/downloads/)). The deterministic run uses only the Python standard library: no package installation, model weights, GPU, or API key is needed.

```bash
git clone https://github.com/rustfuture/rsi-experimental-framework.git
cd rsi-experimental-framework
python -m rsi_framework --config config/default.json --output /tmp/rsi-run
python3 -m rsi_framework --config config/default.json --output /tmp/rsi-run
```

The command prints a JSON summary and writes run artifacts plus `report.md` to `/tmp/rsi-run`, leaving committed results unchanged. On Windows, use your Python 3 command (for example `py -3`) and a local output directory such as `runs/rsi-run` instead of `/tmp/rsi-run`.

For a guided notebook, open [`notebooks/rsi_open_weight_colab.ipynb`](notebooks/rsi_open_weight_colab.ipynb) in Colab. Its default path runs the deterministic harness; the optional local model path needs a GPU runtime.

## How it works
Expand Down
Loading