The developers of Terminal-Bench, a benchmark suite for evaluating the performance of autonomous AI agents on real-world ...
I tested my most powerful magnet yet by sending it to the bottom of a deep lake - and the results were unreal! From strange ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback