Models Model details

Sakana

Sakana: Fugu Ultra

sakana/fugu-ultra

Fugu Ultra is the higher-performance model in Sakana AI's Fugu family. Rather than a single monolithic model, Fugu is a learned multi-agent orchestration system: a language model trained to route tasks across a swappable pool of underlying models and to recursively call instances of itself. Fugu Ultra prioritizes answer quality on complex, multi-step reasoning, coding, and agentic workflows. It supports configurable reasoning effort, tool calling, and built-in web search. Orchestration tokens consumed by the system are billed as standard input/output tokens.

Input
$5

/ 1M tokens

Output
$30

/ 1M tokens

Context window
1M

tokens

Limits

Context

1,000,000

tokens

Max output

128,000

13% of the context window

Model facts

Author

Sakana

Tokenizer

Other

Modality

Text + Image -> Text

Created

Jun 24, 2026

Reasoning

RequiredDefault XhighMaxXhighHigh

Publication status

Hugging Face: not mappedLicense: not publishedParameter count: not published

Modalities

Input · TextInput · ImageOutput · Text

Generation controls

Include reasoningReasoningReasoning effortStructured outputsTool choiceToolsWeb search options