Sakana Fugu

One Model to Command Them All

Frontier-level performance without single-vendor dependency. Fugu dynamically orchestrates the world's best models to tackle complex, multi-step tasks. Plug collective intelligence directly into your workflows today with a single API.

A Multi-Agent System, Delivered as One Model

Sakana Fugu achieves superior performance by dynamically coordinating and orchestrating a diverse pool of powerful models. Instead of using domain knowledge to prescribe team organization, roles, or workflows, Fugu learns to dynamically assemble agents from a pool and coordinate them through non-obvious but highly efficient collaboration patterns.

One API to Access All in an Optimized Way

Access a coordinated pool of specialized models through one API. Fugu handles model selection and switching for each task, reducing API complexity while improving cost-performance.

Offering Superior Performance on Complex Tasks

Built for coding, reasoning, and other quality-critical workflows, Fugu coordinates expert agents to tackle complex tasks with stronger, more reliable results.

Providing Flexibility in Agent Selection

Control which agents can participate in Fugu’s model pool. Opt out of specific providers or models to meet data, privacy, compliance, or organizational requirements.

Research-Driven Coordination for Multi-Agent Intelligence

Sakana Fugu is grounded in two ICLR 2026 papers on learned model orchestration: TRINITY and the Conductor. Together, they show how systems can learn to assemble, route, and coordinate expert agents for each task instead of relying on hand-designed workflows. For a deeper look at the ideas behind the system, explore our technical report.

TRINITY: An Evolved LLM Coordinator

Trinity uses a lightweight evolved coordinator to orchestrate multiple LLMs over several turns, assigning Thinker, Worker, or Verifier roles to adaptively delegate work across coding, math, reasoning, and knowledge tasks.

Learning to Orchestrate Agents in Natural Language with the Conductor

The Conductor is trained with reinforcement learning to discover natural-language coordination strategies, designing agent communication patterns and focused prompts that help diverse LLM pools outperform individual workers on challenging reasoning benchmarks.

Unlock Multi-Agent Intelligence Through An API

Sakana Fugu comes in two models — Fugu and Fugu Ultra — both available through one OpenAI-compatible API. Pick the model that fits your workload, or switch between them without changing your integration.

Fugu

Fugu balances strong performance with low latency, making it the ideal default for everyday work. Drop it into tools like Codex for coding and code review, or power responsive chatbot services — all behind a single endpoint.

Fugu Ultra

Fugu Ultra coordinates a deeper pool of expert agents to maximize answer quality on hard, high-stakes problems. Early users rely on it for Kaggle competitions, paper reproduction, cybersecurity analysis, and literature and patent investigations.

Quantitative Results

Our Fugu models surpass publicly accessible frontier models and are shoulder-to-shoulder with Fable 5 and Mythos Preview in various rigorous engineering, scientific, and reasoning benchmarks while delivering frontier capability without the risk of export controls.

Qualitative Results

These examples compare Sakana Fugu models with three frontier baselines — Gemini 3.1 Pro (high), Opus 4.8 (max), and GPT 5.5 (xhigh).

Example 1 — AutoResearch / LLM学習

AIエージェントに小規模なGPTの学習レシピを自律的に改善させる実験。

Example 2 — 仮名消息の読み順推定

本ケーススタディでは、仮名消息（古典日本語のかな書状）という歴史的資料における読み順の推定問題を対象とする。

Example 3 — ルービックキューブ・ソルバー

ベンチマークでは、Fugu-Ultraと3つのフロンティアモデルそれぞれに、純粋なPythonのみでルービックキューブソルバーをゼロから実装するよう単一のプロンプトを与えた。

Example 4 — CAD メカニカルアイリス

タスク：カメラの絞り（アパーチャ）のような、複数の羽根が連動して動き中央の穴を開閉する機械式アイリスをCADで作成する。

Example 5 — 目隠しチェス

4局の目隠しチェスを連続して対局している。

Example 6 — 株式トレーディング

匿名化された単一銘柄を1つの過去50週間のウィンドウで用いる。

What do our users think about Sakana Fugu?

Software Engineer

For code review, Fugu Ultra is significantly better than GPT-5.5. It gives comprehensive answers and finds the bugs others miss.

Researcher (industry)

I was mapping a patent landscape across ~20 papers and several patents, normally 3–4 days of work. With Fugu I had a full analysis in a few hours.

Executive (enterprise platform)

Raw output quality is on par with top frontier models, but Fugu showed unusually strong persona stability across long sessions.

Researcher

From one simple request, Sakana Fugu worked autonomously for nearly four hours.

Security Engineer

Given one scoped instruction, Sakana Fugu drove a full security assessment end-to-end.

Pricing Plan

Token Plan

Pay-as-you-go for heavy, production workloads needing maximum reliability.

Subscription Plan

We recommend the Token Plan for serious workloads. We offer subscription plans for casual, everyday hands-on use.

FAQ

Q1: How do I get started with Sakana Fugu?
Sakana Fugu is available through an OpenAI-compatible API.

Q2: What's the difference between Fugu and Fugu Ultra?
Fugu balances latency and quality; Fugu Ultra prioritizes accuracy and depth.

Q3: Can I control which models or providers Fugu uses?
Fugu allows opting out of specific models.

Q4: How often do Fugu models update?
We aim to update roughly two weeks after new frontier models are released.

Q5: How does pricing work?
Both subscription and pay-as-you-go plans are available.

Q6: Can I see and predict my token usage and cost?
Yes, token usage and costs are reported per request.

Q7: Will my data be used to train models?
You can opt out of training data usage at any time.

Q8: Can I use Sakana Fugu from outside Japan?
Yes, but we do not provide services to users in EU or EEA member states.