NVIDIA-NeMo / NVIDIA-NeMo/Gym

Performance workstream: agent servers

Open
#3,000 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

agents
Dominant language
Python
Stars
1.2k
Forks
349
Avg merge
1d 21h
Merged PRs (30d)
318

Description

Parent: #2998

Track performance changes in agent rollout loops and agent-specific server setup.

This workstream focuses on repeated request copying and validation, loopback HTTP self-calls, oversized session-seeding payloads, worker scaling, custom app setup, and compatibility with compact token metadata. Changes must preserve session cookies, rollout identity, terminal model-call attribution, telemetry, and the training record returned by /run.

Each implementation or affected agent is tracked as a sub-issue.

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.

Research direction

Start with parent issue #2998 and the sub-issues for each affected agent or implementation; this workstream does not name specific files. Read the rollout loops, agent-specific server setup, and /run path identified by those sub-issues. Done means the scoped performance change preserves cookies, rollout identity, terminal attribution, telemetry, and the returned training record.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
backend, performance
Issue type
Refactor
Difficulty
4/5
Estimated time
3-5 days
Activity status
Active
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.