Sakana Fugu

Not a single model but a learned orchestration system: a language model splits a task, routes the parts to between one and three frontier models and merges the result, served through one OpenAI-compatible API. The base Fugu balances quality against latency.

анонс вендора 22.06.2026. 🔴 сам Sakana не обучал модели-исполнители — это оркестратор, не базовая модель; производителя Sakana AI в справочнике нет

Understands

text, images

Produces

text, code

Contents verified with the vendor: 2026-08-18

Included in subscriptions

No catalogue entry marks this model as part of a plan yet. It will appear here once the plan contents are confirmed with the vendor.