tsbench. The benchmark Token Savior runs at 100% on. 96 real coding tasks, paired runs, oracle-graded ground truth. Plain Claude scores 78.3%; Token Savior scores 192/192. Reproducible with one command.

github.com/Mibayy/tsbench
by @Mibayy

Problem

Audience

Keep exploring

Updates

No founder updates yet. Check back soon.