Skip to content

docs: launch standalone RL trainer check with torchrun #3311

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

Open
CooperSheroy wants to merge 1 commit into
base: main
Choose a base branch
from
Open

docs: launch standalone RL trainer check with torchrun #3311

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
8 changes: 7 additions & 1 deletion README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -151,9 +151,15 @@ uv run sft @ configs/debug/fake/sft.toml
4. Check that you can run the RL trainer (*this requires 1 GPU*)

```bash
uv run trainer @ configs/debug/fake/rl.toml
uv run torchrun \
--standalone \
--nproc-per-node=1 \
-m prime_rl.trainer.rl.train @ configs/debug/fake/rl.toml
```

The standalone RL trainer is still a distributed process group; `torchrun`
sets the rank and rendezvous environment variables expected by PyTorch.

5. Check that you can run the inference server (*this requires 1 GPU*)

```bash
Expand Down