Skip to content

docs(rl): update tutorial with AgenticGRPOLearner for async RL training #4181

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
AntonyMei wants to merge 1 commit into
base: main
Choose a base branch
from
+14 −0
Open

docs(rl): update tutorial with AgenticGRPOLearner for async RL training #4181

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
14 changes: 14 additions & 0 deletions docs/tutorials/posttraining/rl.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -126,6 +126,20 @@ The overview of what this run will do is as follows:
4. Evaluate the policy model's performance on GSM8K math reasoning benchmark
after the post-training with GRPO.

By default, the above command will train the model using GRPOLearner from Tunix. To enable
asynchronous RL training with AgenticGRPOLearner, we can set `rl.use_agentic_rollout` to
true. An example command will be:

```
python3 -m maxtext.trainers.post_train.rl.train_rl \
model_name=${MODEL?} \
load_parameters_path=${MAXTEXT_CKPT_PATH?} \
run_name=${RUN_NAME?} \
base_output_directory=${BASE_OUTPUT_DIRECTORY?} \
chips_per_vm=${CHIPS_PER_VM?} \
rl.use_agentic_rollout=True
```

## Run GSPO

Run the following command for GSPO:
Expand Down
Loading