Sakana AI: Fugu Ultra v2
sakana/fugu-ultra-v2
Access Fugu Ultra v2 from Sakana AI using the Puter.js AI API.
Get Started// npm install @heyputer/puter.js
import { puter } from '@heyputer/puter.js';
puter.ai.chat("Explain quantum computing in simple terms", {
model: "sakana/fugu-ultra-v2"
}).then(response => {
document.body.innerHTML = response.message.content;
});
<html>
<body>
<script src="https://js.puter.com/v2/"></script>
<script>
puter.ai.chat("Explain quantum computing in simple terms", {
model: "sakana/fugu-ultra-v2"
}).then(response => {
document.body.innerHTML = response.message.content;
});
</script>
</body>
</html>
Model Card
Fugu Ultra v2 is Sakana AI's second-generation orchestration model, succeeding the original Fugu Ultra. Rather than a single model, it routes each request across a pool of specialist agents, assigning Thinker, Worker, and Verifier roles behind one OpenAI-compatible endpoint. The approach builds on Sakana's TRINITY and Conductor research from ICLR 2026.
Sakana reports v2 reaches its scores without Fable 5, Fable 5.1, or GPT-6-Astra in its agent pool. On Chartography, a visual and structured-data reasoning benchmark, it scores 48.3 versus Opus 5's 27.3 and Fable 5's 29.5. On the DeepSWE software engineering benchmark it scores 74.3. Sakana ranks it best or joint-best on 5 of 8 published benchmarks.
Pricing matches the original Fugu Ultra, $5 per million input tokens, $30 per million output tokens, with a 1M token context window. It suits complex multi-step reasoning, coding agents, and document or chart analysis, for teams avoiding dependence on one flagship provider.
Context Window 1M
tokens
Max Output 128K
tokens
Input Cost $5
per million tokens
Output Cost $30
per million tokens
Release Date Sep 11, 2026
Model Playground
Try Fugu Ultra v2 instantly in your browser.
This playground uses the Puter.js AI API — no API keys or setup required.
More AI Models From Sakana AI
Fugu Max
Fugu Max is a cost-optimized orchestration model from Tokyo-based Sakana AI, the more affordable sibling to Fugu Ultra v2 in Sakana's Fugu product line. Rather than being one model, it routes each request across the largest pool of open-weight and specialized models Sakana has assembled, including NVIDIA's Nemotron family, under the same TRINITY-based Thinker/Worker/Verifier coordination as Fugu Ultra. Sakana reports Fugu Max takes the best overall score on six benchmarks, including Terminal Bench 2.1, GPQA Diamond, AA-LCR, GDP.pdf, AutomationBench, and its internal SWEFish coding benchmark, and that it extends the cost-performance frontier on seven of ten benchmarks tested. Output pricing is reported as 40-60% lower than Sonnet 5, GPT 5.6 Terra, and Kimi K3. It targets high-volume workloads like coding, document extraction, and multi-step automation where cost per task matters, leaving harder reasoning work to the pricier Fugu Ultra v2.
ChatFugu Ultra
Fugu Ultra is a learned multi-agent orchestration system from Tokyo-based Sakana AI that routes tasks across a swappable pool of frontier LLMs behind a single OpenAI-compatible endpoint. Rather than relying on one monolithic model, Fugu Ultra dynamically assigns Thinker, Worker, and Verifier roles to specialist agents, then synthesizes their outputs into a single response. The underlying coordination is grounded in Sakana's TRINITY and Conductor research, published at ICLR 2026. It targets demanding, multi-step problems: complex reasoning, code review, agentic workflows, cybersecurity analysis, and research tasks. On LiveCodeBench it scores 93.2, ahead of several frontier competitors, and it matches leading models on GPQA-Diamond and Humanity's Last Exam. Fugu Ultra is a strong choice for developers who need frontier-level quality on hard tasks without committing to a single model provider.
Frequently Asked Questions
You can access Fugu Ultra v2 by Sakana AI through Puter.js AI API. Include the library in your web app or Node.js project and start making calls with just a few lines of JavaScript — no backend and no configuration required. You can also use it with Python or cURL via Puter's OpenAI-compatible API.
Fugu Ultra v2 is free to integrate using the Puter.js AI API. With the User-Pays Model, you can add AI to your app for $0, since users cover their own AI usage through their Puter account.
| Price per 1M tokens | |
|---|---|
| Input | $5 |
| Output | $30 |
Fugu Ultra v2 was created by Sakana AI and released on Sep 11, 2026.
Fugu Ultra v2 supports a context window of 1M tokens. For reference, that is roughly equivalent to 2,000 pages of text.
Fugu Ultra v2 can generate up to 128K tokens in a single response.
Yes — the Fugu Ultra v2 API works with any JavaScript framework, Node.js, or plain HTML through Puter.js. Just include the library and start building. See the documentation for more details.
Add Fugu Ultra v2 to your app for free
Developers can integrate Fugu Ultra v2 for free using the Puter.js AI API.
With the User-Pays Model, each user covers their own AI usage instead of the developer.