AWorld: Orchestrating the Training Recipe for Agentic AI

TL;DR

AWorld is an open-source system that significantly accelerates experience generation for training agentic AI, enabling scalable reinforcement learning and achieving state-of-the-art performance on complex benchmarks.

Contribution

We introduce AWorld, a scalable, open-source platform that improves experience collection efficiency and demonstrates its effectiveness by training a high-performing agentic AI model.

Findings

AWorld accelerates experience collection by 14.6x.

The trained agent achieves pass@1 accuracy of 32.23% on GAIA.

The system surpasses prior models like GPT-4o and rivals DeepSeek-V3.

Abstract

The learning from practice paradigm is crucial for developing capable Agentic AI systems, yet it is severely hampered by inefficient experience generation, a bottleneck especially pronounced in complex benchmarks like GAIA. To address this, we introduce AWorld, an open-source system engineered for large-scale agent-environment interaction. By distributing tasks across a cluster, AWorld accelerates experience collection by 14.6x compared to standard single-node, sequential execution. This critical speedup makes extensive reinforcement learning practical and scalable. Leveraging this capability, we trained a Qwen3-32B-based agent that achieves pass@1 accuracy of 32.23% on the GAIA test set, which surpasses GPT-4o (27.91%) and rivals DeepSeek-V3 (31.89%). Our open-source system and the resulting agent provide a practical blueprint for a complete agentic AI training pipeline, from efficient interaction to demonstrable model improvement.