A Japanese model that only delegates work outscored Opus 5.
Sakana AI released two models on Friday that do not answer you, but route your work elsewhere. The stronger one scored 48.3 on a chart-reading test where Anthropic's Opus 5 scored 27.3. Opus 5 is not even in its pool, and the cheaper of the two undercuts frontier prices.
Why it mattersLabs keep selling one giant model, priced as though bigger always wins. A smart dispatcher over cheap models can apparently beat the giant on some work.
Read in the app — free, in 9 languages
Related stories
A Chinese map app built a 3D world you can fly through.2026-09-11An open music model writes the sheet music before the song.
2026-09-11Visa, Mastercard and Ant agreed on how to check an AI shopper.
2026-09-11Cognition built its new coding model on top of a Chinese one.
2026-09-11Russians built a drone swarm with Claude that picks its own targets.
2026-09-11