Vibe Coding DiscoverEvaluate and improve models and agents using environments
Python library for building environments to evaluate and train LLMs and agents. It provides benchmark environments, agent harnesses, verifiers, rollout collection, and scalable execution for evaluation and reinforcement-learning workflows.
Use Cases
Evaluate agents in stateful task environmentsBenchmark models with reproducible verifiersCollect rollouts for reinforcement learningBuild custom training and evaluation environmentsRun large-scale concurrent agent evaluationsInspect and re-verify rollout results
Built With
- Language
- Python
- Frameworks
- vLLM · Ray · LangGraph · NeMo RL · VeRL · Unsloth
Tags
agent-evaluation · LLM-evaluation · reinforcement-learning · benchmarks · training-environments · rollouts · tool-calling · verifiers · agent-training · model-training · scalable-evaluation