NVIDIA-NeMo / NVIDIA-NeMo/Gym

feat: integrate ProRL

Open
#48 0 comments 0 reactions 1 assignee View on GitHub

@joyang-nv is already working on this.

Since Sep 15, 2025.

community-request env-infra
Dominant language
Python
Stars
1.2k
Forks
349
Avg merge
1d 21h
Merged PRs (30d)
318

Description

Use cases, pain points, and background
RL + sandbox, where the LLM interleaves its reasoning process with raw code executions in a sandbox environment. Similar to the ReTool from Bytedance.Especially in the healthcare/life science space. We need sandbox container because of complex dependencies of domain-specific pacakges, and we need to ability for the LLM to write raw code to do sequence preprocessing or post-processing. So a containerized approach like what ProRL built would be great.

Description:

Design:

Out of scope:

Acceptance Criteria:

  • Implementation of the above
  • Test

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.