DukaanBench tests AI agents in a 30-day Indian grocery simulation
is a 30-day of running an Indian grocery store. It is meant to test how well different LLMs can make practical decisions in Indian business settings.
The test covers , customer trust, , perishable goods, and choices made when is limited. The title claims GPT 5.5 performed very well, but the available text does not give scores, comparison models, costs, or token use.
Key points
- simulates 30 days of Indian grocery store .
- It tests LLMs on practical business decisions.
- The tasks include , customer trust, , perishable goods, and limited .
- The title claims GPT 5.5 did very well.
- The excerpt does not include scores, cost data, or .