Which tasks use agentic tool use?

 title: 'OpenAI o3 is the highest performing model on this new benchmark. All models underperform the 80th percentile human score of 36.4%.'

Agentic tool use in the gpt-oss models includes employing various tools to enhance their capabilities. Specifically, the models are trained to use a browsing tool, which allows them to call search functions and interact with the web to fetch information beyond their knowledge cutoff. Additionally, the models can utilize a Python tool, enabling them to execute code in a stateful Jupyter notebook environment. They can also work with arbitrary developer functions, which are specified within the harmony chat format used during training[1].

These capabilities allow the models to interleave tool calls with reasoning processes, improving their performance on tasks that require complex problem-solving. The training emphasizes how to effectively incorporate these tools while managing user interactions[1].

Space: Let’s explore the gpt-oss-120b and gpt-oss-20b Model Card