> For clean Markdown of any page, append .md to the page URL. > For a complete documentation index, see https://docs.nvidia.com/nemo/gym/llms.txt. > For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/nemo/gym/_mcp/server. # Ecosystem > How NeMo Gym fits within the NVIDIA NeMo platform and the broader agentic ecosystem. NeMo Gym is one of several specialized libraries in the NVIDIA NeMo platform. Alongside NeMo RL for post-training, NeMo Gym focuses on the *environments* agents act in — pairing a dataset, agent harness, and verifier so the same environment can drive both evaluation and training. For what NeMo Gym does and when to use it, see the [Overview](/about) and [What NeMo Gym Provides](/about#what-nemo-gym-provides). For how environments are implemented as composable servers, see [Architecture](/about/architecture). NeMo Gym is also integrated with the broader agentic ecosystem. We would love your contribution! Open a PR to add an integration, or [file an issue](https://github.com/NVIDIA-NeMo/Gym/issues/new/choose) to share what would be valuable for you. ## NeMo Gym Within NVIDIA NeMo NVIDIA NeMo is a GPU-accelerated platform for training generative AI models and optimizing AI agents. It includes specialized libraries that span the model lifecycle end to end: * **NeMo Curator**: Prepares, filters, and deduplicates training data at scale. * **NeMo Gym**: Defines the environments — dataset, agent harness, and verifier — that agents are evaluated and trained in *(this library)*. * **NeMo RL**: Runs scalable post-training (GRPO, DPO, and SFT) against those environments. **NeMo Gym's role**: NeMo Gym is where a task becomes an interactive environment. It standardizes how an agent interacts with the world (the agent harness), how task completion is scored (the verifier), and how per-task state is isolated — producing the rollouts and reward signals that training frameworks like NeMo RL consume, and the consistent scoring that evaluation depends on. ## Environment Libraries Seamlessly combine environments and benchmarks from other libraries alongside NeMo Gym environments, with access to over 1,000 community-contributed environments across integrated libraries. * **[Aviary](https://github.com/NVIDIA-NeMo/Gym/tree/main/resources_servers/aviary)** * **[Harbor](https://github.com/NVIDIA-NeMo/Gym/tree/main/responses_api_agents/harbor_agent)** * **[OpenEnv](https://github.com/NVIDIA-NeMo/Gym/tree/main/resources_servers/openenv)** * **[Reasoning Gym](https://github.com/NVIDIA-NeMo/Gym/tree/main/resources_servers/reasoning_gym)** * **[Verifiers](https://github.com/NVIDIA-NeMo/Gym/tree/main/responses_api_agents/verifiers_agent)** ## Training Framework Libraries Use environments for SFT and RL training. If you're interested in integrating another training framework, see the [Training Framework Integration Guide](/contribute/rl-framework-integration). * **[NeMo RL](/training-tutorials/nemo-rl-grpo)** * **[Unsloth](/training-tutorials/unsloth)** * **[VeRL](/training-tutorials/verl)** ## Agent Harnesses Agent harnesses are available out of the box for evaluation and training. Some examples: * **[OpenHands](https://github.com/NVIDIA-NeMo/Gym/tree/main/responses_api_agents/swe_agents)** - software engineering agent harness * **[Mini SWE Agent](https://github.com/NVIDIA-NeMo/Gym/tree/main/responses_api_agents/mini_swe_agent)** - software engineering agent harness * **[LangGraph](https://github.com/NVIDIA-NeMo/Gym/tree/main/responses_api_agents/langgraph_agent)** - agent patterns built with LangGraph (reflection, orchestration, parallel thinking) ## Related Products in the NVIDIA NeMo Ecosystem Depending on your workflow, you may also find these NeMo libraries useful across data preparation, evaluation and training, and deployment. ### Data | Library | Purpose | | ------------------------------------------------------------------------ | -------------------------------------------------------------- | | [NeMo Curator](https://github.com/NVIDIA-NeMo/Curator) | Scalable data preprocessing and curation | | [NeMo Data Designer](https://github.com/NVIDIA-NeMo/DataDesigner) | Generate synthetic training data from scratch or seed examples | | [NeMo Safe Synthesizer](https://github.com/NVIDIA-NeMo/Safe-Synthesizer) | Generate privacy-safe synthetic copies of sensitive datasets | | [NeMo Anonymizer](https://github.com/NVIDIA-NeMo/Anonymizer) | Detect and replace sensitive data | ### Evaluation & Training | Library | Purpose | | ---------------------------------------------------------------------- | -------------------------------------------------------------------------- | | **[NeMo Gym](https://github.com/NVIDIA-NeMo/Gym)** | Evaluate and improve models and agents using environments *(this project)* | | [NeMo Evaluator](https://github.com/NVIDIA-NeMo/Evaluator) | Model evaluation and benchmarking | | [NeMo RL](https://github.com/NVIDIA-NeMo/RL) | Scalable post-training with GRPO, DPO, and SFT | | [NeMo Megatron-Bridge](https://github.com/NVIDIA-NeMo/Megatron-Bridge) | Pretraining and fine-tuning with Megatron-Core | | [NeMo AutoModel](https://github.com/NVIDIA-NeMo/Automodel) | PyTorch native training for Hugging Face models | ### Deployment | Library | Purpose | | ------------------------------------------------------------------ | --------------------------------------------------- | | [NeMo Agent Toolkit](https://github.com/NVIDIA/NeMo-Agent-Toolkit) | Connect and optimize teams of AI agents | | [NeMo Guardrails](https://github.com/NVIDIA-NeMo/Guardrails) | Enforce safety and policy rules | | [NeMo Retriever](https://github.com/NVIDIA/NeMo-Retriever) | Document extraction and retrieval for RAG pipelines |