AWorld: Orchestrating the Training Recipe for Agentic AI
By: Chengyue Yu , Siyuan Lu , Chenyi Zhuang and more
Potential Business Impact:
AI learns much faster to solve hard problems.
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.
Similar Papers
AWorld: Orchestrating the Training Recipe for Agentic AI
Artificial Intelligence
AI learns faster, solving harder puzzles.
Benchmarking World-Model Learning
Artificial Intelligence
Tests AI learning better for many tasks.
Benchmarking World-Model Learning
Artificial Intelligence
Tests how well AI learns about the world.