---
id: "sakana/fugu-ultra"
also: ["sakana/fugu-ultra-20260615", "requesty/sakana/fugu-ultra", "vercel/sakana/fugu-ultra"]
card_usd_per_m: "5.0 in, 30.0 out, at sakana"
cheapest_list_usd_per_m: "5.0 in, 30.0 out, at sakana on openrouter"
cheapest_effective_usd_per_m: "2.356 in, 30.0 out, at sakana on requesty"
context: 1000000
max_output: 128000
input: ["text", "image"]
output: ["text"]
released: "2026-06-24"
reasoning: ["max", "xhigh", "high"]
api: ["/api/v1/chat/completions", "/api/v1/completions", "/api/v1/messages", "/api/v1/responses"]
params: ["include_reasoning", "max_tokens", "reasoning", "reasoning_effort", "response_format", "stop", "structured_outputs", "temperature", "tool_choice", "tools", "web_search_options"]
endpoints: 3
gateways: "openrouter (1), requesty (1), vercel (1)"
no_training: 3
trains: 0
retains_prompts: 3
---
# Sakana: Fugu Ultra

Fugu Ultra is built on a pool of publicly accessible frontier models, rather than running as a single model. It coordinates several models, routing work to 1-3 agents depending on the problem and combining their results into a single answer.