Fugu review, pricing and limits

Fugu suits engineering and research teams who want frontier-level output without locking into one model vendor: it assigns whichever specialist model fits a step and checks shaky answers before returning a result. It supports a 272K-token context window before higher rates kick in. It is not a fit if you need to know exactly which model touched a given request.

Fugu is Sakana AI's fully autonomous multi-agent orchestrator, launched June 2026. A trained conductor model assigns each task to specialist agents from its pool, then verifies the result before returning one synthesized answer. Fugu Ultra reaches 95.5% on GPQA Diamond, Sakana's highest published reasoning score for the system.

Maker: Sakana AI · Autonomy: fully autonomous · Maturity: GA

Underlying models: Sakana Conductor (proprietary orchestrator, trained via reinforcement learning), Pool of closed and open partner models (not fully disclosed by Sakana AI)

About Fugu

Fugu is Sakana AI's multi-agent orchestration system, delivered as a single OpenAI-compatible API endpoint rather than one monolithic model. Instead of answering every prompt itself, Fugu is a trained conductor model that decides which specialist models to call from an agent pool, whether to split a task into planning and execution steps, when to have another agent verify the result, and how to synthesize the final answer. Sakana AI, the Tokyo-based lab behind it (founded 2023), built Fugu on two internal research frameworks: TRINITY, which explores evolutionary coordination across Thinker, Worker, and Verifier roles, and Conductor, which uses reinforcement learning to discover collaboration strategies between agents rather than hand-wiring them. The lab raised a $135M Series B in November 2025 at roughly a $2.65B valuation, led by Mitsubishi UFJ, Lux Capital, and In-Q-Tel, funding its shift from research demos toward production systems like Fugu.

Fugu ships in two tiers. The base model is tuned for low latency and interactive use, acting as a default engine for chat and coding assistants. Fugu Ultra trades speed for quality, drawing on a larger pool of expert agents for complex, high-stakes work such as automated ML research, cybersecurity analysis, and multi-step patent investigation. What makes this an agent rather than a chatbot is that the orchestration is learned end-to-end: Fugu can recursively call itself and other agents without a human approving each intermediate step, deciding on its own when to delegate, when to verify, and when a first answer is shaky enough to warrant a second opinion.

On benchmarks, Fugu Ultra scores 73.7% on SWE-bench Pro, 93.2% on LiveCodeBench v6, 95.5% on GPQA Diamond, 50.0% on Humanity's Last Exam, and 82.1% on TerminalBench 2.1, according to Sakana's June 2026 release materials. Sakana positions these scores as ahead of Claude Opus 4.8 (69.2% SWE-bench Pro), GPT-5.5 (58.6%), and Gemini 3.1 Pro (54.2%) on the same task, and roughly matching Anthropic's Fable 5 and Mythos Preview despite Fugu not being a single frontier-scale model itself. These figures are Sakana-reported and have not been independently reproduced on public leaderboards.

Fugu launched publicly on June 22, 2026, priced as a subscription with a separate metered API (see pricing for exact figures). The API matches the OpenAI Chat Completions format, so teams can swap an existing endpoint URL and get orchestrated multi-agent behavior without rearchitecting their application. Standard-tier users can opt out of specific underlying providers; the Ultra tier uses a fixed agent pool with no opt-out. Sakana frames Fugu's core pitch as reduced single-vendor lock-in: by routing across multiple providers behind one endpoint, teams get insulated from any single vendor's outages, export-control changes, or policy shifts. An open-source alternative called Maestro solves the same orchestration problem from the opposite direction, self-hosted and fully transparent, versus Fugu's closed and centrally trained approach.

Pricing

Three subscription tiers at $20, $100, and $200 per month. Standalone API bills per token at roughly $5 per 1M input tokens and $30 per 1M output tokens, with higher rates above 272K tokens of context. No confirmed free tier as of the June 22, 2026 launch.

Key Features

• Learned Orchestration: A trained conductor model decides which specialist agents handle a task and how to combine their outputs, instead of a hand-coded routing script.
• Recursive Self-Calling: Fugu can call itself and other agents repeatedly for test-time scaling on hard problems, without retraining or a human approving each step.
• Two-Tier Model Line: Base Fugu targets low-latency interactive use; Fugu Ultra draws on a larger expert-agent pool for high-stakes tasks like ML research and cybersecurity analysis.
• OpenAI-Compatible Endpoint: One API endpoint matches the OpenAI Chat Completions format, so existing GPT, Gemini, or Claude integrations can switch over with minimal code changes.
• Built-In Verification Step: A shaky answer can be routed to a second agent for review before Fugu returns a final result, instead of always returning the first agent's output.

Strengths

• Sakana reports Fugu Ultra leading named single-model competitors on its own coding and reasoning benchmark comparison, not just matching them.
• The OpenAI-compatible API lets teams already built on GPT, Gemini, or Claude endpoints adopt orchestration by swapping a URL rather than rewriting integration code.
• Routing across multiple underlying providers behind one endpoint reduces exposure to any single vendor's outages or policy changes, by design.

Weaknesses

• Not available in the EU or EEA as of the June 2026 launch, while Sakana completes GDPR compliance work.
• Sakana has not comprehensively disclosed which models sit in Fugu's agent pool, making it hard to audit exactly what is processing a given request.
• Output tokens cost more per million than several single-model frontier APIs, and a complex orchestrated task can burn through more output tokens than a single-model call would.

Visit Fugu Official Site