Benchmark

TextCraft

A text-based agent planning benchmark used to evaluate language models' multi-step planning and execution capabilities in structured tasks.

Timeline (last 90 days)

Oct 9

After targeted training using the ALFWorld and TextCraft benchmarks, the two types of planners performed similarly on plan generation success rate.

Unverified50%

After targeted training using the ALFWorld and TextCraft benchmarks, the two types of planners performed similarly on plan generation success rate.