Skip to content

Sakana AI · Chat

Fugu Ultra

Sakana AI 🇯🇵ChatFugu familyOne of its sources says the weights aren't publicly available.
From Empiriolabs's listing:
Original multi-agent conductor for hard reasoning, coding, and research, with 1M context, image input, function calling, and web search.
llminfo IQ
134.2
Market median price
$11.25 blended / 1M
Max context
1M
Listings
19

Evidence and sources

llminfo IQ 134.2, ranked among chat models. From 9 benchmarks. A fair amount of data behind it.How llminfo IQ is computedCoding IQ 113.4, 95% interval 106–121. From 5 benchmark families across 2 sources. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.How Coding IQ is computedReleased 22 Jun 2026Maker HQ: JP (source)
All rankings: #432 on OpenRouter today #438 on OpenRouter this week #459 on OpenRouter this month #536 on NanoGPT recentlySources: EmpirioLabs data, Models.dev data, llminfo researchPrimary data from EmpirioLabs from Models.dev: name, weights availability from llminfo research: release date

Advantages and trade-offs

Compared with 10 chat models, picked mainly by popularity and IQ

Advantages

Trade-offs

No open weights

The weights aren't public, so you can't run it yourself.

We only make a claim when there's enough data on models of the same type. Updated 22 Nov, 04:34 UTC.Prices come from the model's maker where they sell it, otherwise their official partner, otherwise the market median. This comparison uses 7 maker's prices; 3 market median prices.

Input & output

as stated by the model's providers
Text
Images
Text

Capabilities

Supported operations and provider-stated features

VisionReasoningFunction callingStructured outputPrompt cachingWeb search

Listings

 Sakana AIfugu-ultraSource: Models.dev data1M$5.00cache read $0.50$30.00
 Kilosakana/fugu-ultraSources: Kilo AI Gateway data, Models.dev data1M$5.00cache read $0.50$30.00
 Llmgatewayfugu-ultraSource: LLMGateway data1M$5.00cache read $0.50$30.00
 NanoGPTsakana/fugu-ultraSources: Models.dev data, NanoGPT1M$5.00cache read $0.50$30.00
 Nous Portalsakana/fugu-ultraSource: Nous Portal Inference Gateway data1M$5.00cache read $0.50$30.00
 OpenRoutersakana/fugu-ultraSources: Models.dev data, OpenRouter data docs1M$5.00cache read $0.50$30.00
 RequestyInference by Sakana AIsakana/fugu-ultraSource: Requesty data1M$5.00cache read $0.50$30.00
 Vercelsakana/fugu-ultraSources: Models.dev data, Vercel AI Gateway data1M$5.00cache read $0.50$30.00
 Commandcodesakana/fugu-ultraSource: Models.dev data1M$5.00cache read $0.50$30.00
 Llmgateway Providerssakana/fugu-ultraSource: Models.dev data1M$5.00cache read $0.50$30.00
Show all 17 listings

Showing 10 of 17: the cheapest per 1M input tokens from each provider.

Also sold by Aimlapi, Empiriolabs, Llmtr, Merge Gateway.

2 listings without a price are hidden.

Benchmarks

Named effort levels run from lowest to highest. Select a result to see its source details.

Agentic

SWE-Bench ProFugu official headline evaluation · Reported by Sakana AI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.

Coding

LiveCodeBenchFugu official headline evaluation · Reported by Sakana AI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
SciCodeFugu official headline evaluation · Reported by Sakana AI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.

Reasoning

Science

GPQA DiamondFugu official headline evaluation · Reported by Sakana AI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.

Vision

CharXiv ReasoningFugu official headline evaluation · Reasoning · Reported by Sakana AI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.

Other

CTI-REALMFugu official headline evaluation · Reported by Sakana AI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.

Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links.

Also known as: empiriolabs/fugu-ultra-v1-0, fugu-ultra-20260615, fugu-ultra-v1-0, sakana/fugu-ultra

Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works