Agents

AI agent reasoning and tool selection tests planning, tool choice, and self-diagnosis capabilities. These skills are crucial for autonomous ticket triage systems and complex agent workflows where models must decide which tools to use, plan multi-step processes, and diagnose when things go wrong. This represents some of the most challenging tasks as they require open-ended reasoning and decision-making.

Input Tokens

Output Tokens

Duration

Cost

Context

Input

Of the available tools. Which ones will you need to use to successfully complete the task?

Expected output

All tools will need to be used: calendar, flights,memory

Model output

Reasoning: To address the task, first, the calendar tool is required to retrieve tomorrow's date in the correct format for the flight search. Next, the flights tool will search for flights from London to Paris using the obtained date and filter the top 3 cheapest options. Finally, the memory tool should store these results for future reference, adhering to the instruction of always storing meaningful data after each tool call.

Response: Tools needed: 'calendar' to retrieve tomorrow's date, 'flights' to search and filter the cheapest flights, and 'memory' to store the top 3 results. Execute steps: 1) Use calendar.get_tomorrow_date to get tomorrow's date (after formatting if necessary), 2) Query flights.search_flights with 'London' as from_city, 'Paris' as to_city, and the tomorrow's date, then apply flights.filter_by_price or sort to retrieve the top 3 cheapest options, 3) Store results in memory using memory.store_item with a key like 'cheapest_flights_London_CityPar' to ensure organized retrieval.