Skip to content

Sakana Fugu Ultra v2

Sakana AI · Released 11 Sep 2026 · fugu-ultra-v2.0

Compare this model
ProprietaryReasoningCodingResearchAgents

$5.00

Input price / 1M tokens

#52 of 61, cheapest first

$30.00

Output price / 1M tokens

#57 of 61, cheapest first

–

Context window

–

Output tokens / second

74.3%

DeepSWE v1.1

#2 of 17

Summary

Sakana Fugu Ultra v2 is a proprietary model from Sakana AI, released on 11 Sep 2026.

At $5.00 input and $30.00 output per million tokens, it is #52 of 61 on input price, cheapest first.

A multi-agent orchestration system delivered as one model API: it coordinates a fixed pool of open and specialized models to maximise answer quality on hard, multi-step reasoning, research and software tasks.

Benchmarks · 2 reported

BenchmarkScoreBarRankSource
Chartography48.3%–Sakana launch postSelf-reported
DeepSWE v1.1version not stated74.3%#2 of 17Sakana launch postSelf-reported

Self-reported means the lab ran the test itself. A rank appears only where several labs report the same version of a benchmark.

Similar models

Models with the most benchmarks in common with Sakana Fugu Ultra v2, and the closest scores.

Compare 4 side by side

DeepSWE v1.1

#2 of 17

Long coding tasks in real repos

Compare Sakana Fugu Ultra v2 side by side

Pre-filled with the two closest models. Swap any of them.

Open comparison

Sources

  • price, long-context tier above 272K, cached price, API ID fugu-ultra-v2.0, plans, EU/EEA restriction: sakana.ai/fugu/, checked 2 Oct 2026
  • release date (11 Sep 2026) and benchmarks: sakana.ai/fugu-max-release/, checked 2 Oct 2026

Building on Sakana Fugu Ultra v2?

Get the architecture right before the bill arrives.

We size caching, routing and fallbacks for your workload on a free call, and tell you where a cheaper model is good enough.