NeMo Gym

Requirements • Quick Start • Available Environments • Documentation & Resources • Community & Support • Citations

NeMo Gym is a library for building reinforcement learning (RL) training environments for large language models (LLMs). It provides infrastructure to develop environments, scale rollout collection, and integrate seamlessly with your preferred training framework.

🏆 Why NeMo Gym?

• Scaffolding and patterns to accelerate environment development: multi-step, multi-turn, and user modeling scenarios
• Contribute environments without expert knowledge of the entire RL training loop
• Test environments and throughput end-to-end, independent of the RL training loop
• Interoperable with existing environments, systems, and RL training frameworks
• Growing collection of training environments and datasets for Reinforcement Learning from Verifiable Reward (RLVR)

Important

NeMo Gym is currently in early development. You should expect evolving APIs, incomplete documentation, and occasional bugs. We welcome contributions and feedback - for any changes, please open an issue first to kick off discussion!

🔗 Ecosystem

NeMo Gym is part of NVIDIA NeMo, NVIDIA's GPU-accelerated platform for building and training generative AI models. NeMo Gym integrates with a growing number of RL training frameworks and environment libraries; see the Ecosystem page for full details and tutorials.

Training Frameworks: NeMo RL • OpenRLHF • TRL • Unsloth • more →

Environment Libraries: Reasoning Gym • Aviary • more →

📋 Requirements

NeMo Gym is designed to run on standard development machines:

Hardware Requirements	Software Requirements
GPU: Not required for NeMo Gym library operation • GPU may be needed for specific resource servers or model inference (see individual server documentation)	Operating System: • Linux (Ubuntu 20.04+, or equivalent) • macOS (11.0+ for x86_64, 12.0+ for Apple Silicon) • Windows (via WSL2)
CPU: Any modern x86_64 or ARM64 processor (e.g., Intel, AMD, Apple Silicon)	Python: 3.12 or higher
RAM: Minimum 8 GB (16 GB+ recommended for larger environments)	Git: For cloning the repository
Storage: Minimum 5 GB free disk space for installation and basic usage	Internet Connection: Required for downloading dependencies and API access

Additional Requirements

• API Keys: OpenAI API key with available credits (for the quickstart examples)
  • Other model providers supported (Azure OpenAI, self-hosted models via vLLM)
• Ray: Automatically installed as a dependency (no separate setup required)

🚀 Quick Start

Install NeMo Gym, start the servers, and collect your first verified rollouts for RL training.

Setup

# Clone the repository
git clone git@github.com:NVIDIA-NeMo/Gym.git
cd Gym

# Install UV (Python package manager)
curl -LsSf https://astral.sh/uv/install.sh | sh
source $HOME/.local/bin/env

# Create virtual environment
uv venv --python 3.12
source .venv/bin/activate

# Install NeMo Gym
uv sync --extra dev --group docs

Configure Your API Key

Create an env.yaml file that contains your OpenAI API key and the policy model you want to use. Replace your-openai-api-key with your actual key. This file helps keep your secrets out of version control while still making them available to NeMo Gym.

echo "policy_base_url: https://api.openai.com/v1
policy_api_key: your-openai-api-key
policy_model_name: gpt-4.1-2025-04-14" > env.yaml

Note

We use GPT-4.1 in this quickstart because it provides low latency (no reasoning step) and works reliably out-of-the-box. NeMo Gym is not limited to OpenAI models—you can use self-hosted models via vLLM or any OpenAI-compatible inference server. See the documentation for details.

Start Servers

Terminal 1 (start servers):

# Start servers (this will keep running)
config_paths="resources_servers/example_single_tool_call/configs/example_single_tool_call.yaml,\
responses_api_models/openai_model/configs/openai_model.yaml"
ng_run "+config_paths=[${config_paths}]"

Terminal 2 (interact with agent):

# In a NEW terminal, activate environment
source .venv/bin/activate

# Interact with your agent
python responses_api_agents/simple_agent/client.py

Collect Rollouts

Terminal 2 (keep servers running in Terminal 1):

# Create a simple dataset with one query
echo '{"responses_create_params":{"input":[{"role":"developer","content":"You are a helpful assistant."},{"role":"user","content":"What is the weather in Seattle?"}]}}' > weather_query.jsonl

# Collect verified rollouts
ng_collect_rollouts \
    +agent_name=example_single_tool_call_simple_agent \
    +input_jsonl_fpath=weather_query.jsonl \
    +output_jsonl_fpath=weather_rollouts.jsonl

# View the result
cat weather_rollouts.jsonl | python -m json.tool

This generates training data with verification scores!

Clean Up Servers

Terminal 1 with the running servers: Ctrl+C to stop the ng_run process.

Next Steps

Now that you can generate rollouts, choose your path:

• Start training — Train models using NeMo Gym with your preferred RL framework. See the Training Tutorials.

• Use an existing environment — Browse the Available Environments below to find an environment that matches your goals.

• Build a custom environment — Implement or integrate existing tools and define task verification logic. Get started with the Creating a Training Environment tutorial.

📦 Available Environments

NeMo Gym includes a curated collection of environments for training and evaluation across multiple domains:

Example Environment Patterns

Purpose: Demonstrate NeMo Gym patterns and concepts.

Name	Demonstrates	Config	README
Multi Step	Multi-step tool calling	example_multi_step.yaml	README
Session State Mgmt	Session state management (in-memory)	example_session_state_mgmt.yaml	README
Single Tool Call	Basic single-step tool calling	example_single_tool_call.yaml	README

Environments for Training & Evaluation

Purpose: Training-ready environments with curated datasets.

Tip

Each resource server includes example data, configuration files, and tests. See each server's README for details.

Resource Server	Config	Domain	Dataset	Description	Value	Train	Validation	License
Arc Agi	arc_agi.yaml	knowledge	-	-	-	-	✓	-
Aviary	aviary.yaml	math	-	-	-	✓	✓	Apache 2.0
Aviary	bixbench_aviary.yaml	coding	-	-	-	-	-	-
Aviary	gsm8k_aviary.yaml	math	-	-	-	✓	✓	Apache 2.0
Aviary	hotpotqa_aviary.yaml	agent	-	-	-	✓	✓	Apache 2.0
Calendar	calendar.yaml	agent	Nemotron-RL-agent-calendar_scheduling	-	-	✓	✓	Apache 2.0
Code Gen	code_gen.yaml	coding	nemotron-RL-coding-competitive_coding	-	-	✓	✓	Apache 2.0
Equivalence Llm Judge	equivalence_llm_judge.yaml	knowledge	-	Short answer questions with LLM-as-a-judge	Improve knowledge-related benchmarks like GPQA / HLE	-	-	-
Equivalence Llm Judge	lc.yaml	knowledge	-	-	-	-	-	-
Equivalence Llm Judge	lc_judge.yaml	knowledge	-	-	-	-	-	-
Equivalence Llm Judge	nl2bash-equivalency.yaml	agent	-	Short bash command generation questions with LLM-as-a-judge	Improve foundational bash and IF capabilities	✓	✓	GNU General Public License v3.0
Google Search	google_search.yaml	agent	Nemotron-RL-knowledge-web_search-mcqa	Multi-choice question answering problems with search tools integrated	Improve knowledge-related benchmarks with search tools	✓	-	Apache 2.0
Instruction Following	instruction_following.yaml	instruction_following	Nemotron-RL-instruction_following	Instruction following datasets targeting IFEval and IFBench style instruction following capabilities	Improve IFEval and IFBench	✓	-	Apache 2.0
Math Advanced Calculations	math_advanced_calculations.yaml	agent	Nemotron-RL-math-advanced_calculations	An instruction following math environment with counter-intuitive calculators	Improve instruction following capabilities in specific math environments	✓	-	Apache 2.0
Math Formal Lean	math_formal_lean.yaml	math	-	Lean4 formal proof verification environment	Improve formal theorem proving capabilities	✓	-	MIT
Math Formal Lean	math_formal_lean_multi_turn.yaml	math	-	Lean4 formal proof verification environment with multi-turn self-correction	Improve formal theorem proving capabilities	✓	-	MIT
Math Formal Lean	nemotron_clean_easy.yaml	math	-	Lean4 formal proof verification environment	Improve formal theorem proving capabilities	✓	-	Apache 2.0
Math Formal Lean	nemotron_first_try_hard.yaml	math	-	Lean4 formal proof verification environment	Improve formal theorem proving capabilities	✓	-	Apache 2.0
Math Formal Lean	nemotron_medium_500.yaml	math	-	Lean4 formal proof verification environment	Improve formal theorem proving capabilities	✓	-	Apache 2.0
Math Formal Lean	nemotron_very_easy.yaml	math	-	Lean4 formal proof verification environment	Improve formal theorem proving capabilities	✓	-	Apache 2.0
Math With Code	math_with_code.yaml	math	-	-	-	✓	-	Apache 2.0
Math With Judge	bytedtsinghua_dapo17k.yaml	math	-	-	-	✓	✓	Apache 2.0
Math With Judge	dapo17k.yaml	math	-	-	-	✓	✓	Apache 2.0
Math With Judge	dapo17k_filtered_qwen330ba3binstruct.yaml	math	-	-	-	✓	✓	Apache 2.0
Math With Judge	dapo17k_trajectory_collection.yaml	math	-	-	-	-	✓	-
Math With Judge	math_stack_overflow.yaml	math	Nemotron-RL-math-stack_overflow	-	-	✓	✓	Creative Commons Attribution-ShareAlike 4.0 International
Math With Judge	math_with_judge.yaml	math	Nemotron-RL-math-OpenMathReasoning	Math dataset with math-verify and LLM-as-a-judge	Improve math capabilities including AIME 24 / 25	✓	✓	Creative Commons Attribution 4.0 International
Math With Judge	math_with_local_judge.yaml	math	-	-	-	-	-	-
Mcqa	mcqa.yaml	knowledge	Nemotron-RL-knowledge-mcqa	Multi-choice question answering problems	Improve benchmarks like MMLU / GPQA / HLE	✓	-	Apache 2.0
Mini Swe Agent	mini_swe_agent.yaml	coding	SWE-Gym	A software development with mini-swe-agent orchestration	Improve software development capabilities, like SWE-bench	✓	✓	MIT
Multichallenge	multichallenge.yaml	knowledge	-	MultiChallenge benchmark evaluation with LLM judge	-	✓	-	TBD
Multichallenge	multichallenge_nrl.yaml	knowledge	-	MultiChallenge benchmark evaluation with LLM judge	-	✓	-	TBD
Newton Bench	newton_bench.yaml	math	-	-	-	✓	-	Apache 2.0
Ns Tools	ns_tools.yaml	agent	-	NeMo Skills tool execution with math verification	-	-	-	-
Reasoning Gym	reasoning_gym.yaml	knowledge	-	-	-	✓	-	Apache 2.0
Reasoning Gym	resources_only.yaml	knowledge	-	-	-	-	-	-
Structured Outputs	structured_outputs_json.yaml	instruction_following	Nemotron-RL-instruction_following-structured_outputs	Check if responses are following structured output requirements in prompts	Improve instruction following capabilities	✓	✓	Apache 2.0
Swerl Gen	swerl_gen.yaml	coding	-	Running sandboxed evaluation for SWE-style tasks (either patch generation or reproduction test generation)	Improve SWE capabilities useful for benchmarks like SWE-bench	✓	✓	Apache 2.0
Swerl Llm Judge	swerl_llm_judge.yaml	coding	-	SWE-style multiple-choice LLM-judge tasks scored via ... choice.	Improve SWE capabilities useful for benchmarks like SWE-bench	✓	✓	MIT
Terminus Judge	terminus_judge.yaml	agent	-	single-step terminal based task	Improve on terminal-style tasks	✓	✓	Apache 2.0
Text To Sql	text_to_sql.yaml	coding	-	Text-to-SQL generation with LLM-as-a-judge equivalence checking	Improve text-to-SQL capabilities across multiple dialects	-	-	-
Workplace Assistant	workplace_assistant.yaml	agent	Nemotron-RL-agent-workplace_assistant	Workplace assistant multi-step tool-using environment	Improve multi-step tool use capability	✓	✓	Apache 2.0
Xlam Fc	xlam_fc.yaml	agent	-	-	-	✓	✓	Apache 2.0

📖 Documentation & Resources

• Documentation - Technical reference docs
• Training Tutorials - Train with NeMo Gym environments
• API Reference - Complete class and function reference

🤝 Community & Support

We'd love your contributions! Here's how to get involved:

• Report Issues - Bug reports and feature requests
• Contributing Guide - How to contribute code, docs, new environments, or training framework integrations

📚 Citations

If you use NeMo Gym in your research, please cite it using the following BibTeX entry:

@misc{nemo-gym,
  title = {NeMo Gym: An Open Source Library for Scaling Reinforcement Learning Environments for LLM},
  howpublished = {\url{https://github.com/NVIDIA-NeMo/Gym}},
  author={NVIDIA},
  year = {2025},
  note = {GitHub repository},
}