sakana

Fugu Max is the largest tier of Sakana AI's Fugu line — a 1M-token-context model that takes text, images, and files and returns text. It joins sakana/fugu and sakana/fugu-ultra as a distinct tier, and is not a v2 of fugu-ultra (that one is folded into the existing row as an alias, not a new listing).

Fugu Ultra is Sakana AI's flagship orchestration model. Rather than answering directly, it coordinates a deep pool of expert agents and frontier models behind one OpenAI-compatible API, decomposing a task, dispatching sub-problems to the best available model, and synthesizing the result. It supports OpenAI-compatible Chat Completions and Responses APIs, including built-in tools such as web_search, and targets long-horizon, agentic, and research-grade workloads with a large context window.

Fugu is Sakana AI's orchestration model, balancing strong performance with low latency. Rather than answering directly, it routes a task across a pool of expert agents and frontier models behind one OpenAI-compatible API, then synthesizes the result. Platform usage bills at the standard rate of the underlying model that Fugu selects, so effective cost varies per request. It supports OpenAI-compatible Chat Completions and Responses APIs, including built-in tools such as web_search.

Sakana Namazu is Sakana AI's Japanese-specialized LLM, built on the open model Kimi K2.6 and tuned in-house for Japanese and Japanese business contexts. It is a single in-house model (unlike Fugu's orchestrator) offered via an OpenAI-compatible API, and combines web search and code execution to carry complex tasks through to completion.