Skip to contentT/TTokens or Towers

Compare

LeaderboardEvery tracked model, ranked and priced side by sideBenchmarksWhat each test measures and who leads itPricesAPI, media, and hardware rates with a cost calculator

Plan

Stack plannerSix questions to a priced local and hosted stackShortlistOur picks for subscriptions, APIs, and hardwareEnterpriseSeat plans against self-hosting for a whole team

Build

HardwareGPUs and workstations, and which models they fitHostsWhere to rent open-weight inference, and for how muchHarnessesCoding agents and the models they drive

Lab

FindingsResults from our own evaluationsScanAudit a repo’s AI usage in the browserMethodWhere the numbers come from and how we check them
Data checked September 25, 2026
Data checked September 25, 2026
Plan a stack
T/TTokens or Towers

An independent guide to AI models, what they cost, and the hardware to run them. No account, no API key, no sponsors.

Compare

  • Leaderboard
  • Benchmarks
  • Prices

Plan

  • Stack planner
  • Shortlist
  • Enterprise

Build

  • Hardware
  • Hosts
  • Harnesses

Lab

  • Findings
  • Scan
  • Method
Data checked September 25, 2026 · Every number links to its sourceRSS · Source on GitHub
Leaderboard/Sakana AI
Sakana AI

Sakana AI · Fugu

Fugu Ultra v2

Closed weightsGenerally availableNewproprietary

Sakana AI's top hosted multi-model orchestration system (routes and combines frontier models from its pool), aimed at hard coding, research and document tasks via API.

AA Intelligence
—
Input / output per 1M
$5.00/ $30.00
Output speed
—
Context window
—

Reasoning and knowledge

  • Humanity’s Last Exam#3 of 147
    56.3%vendortext onlySakana AI: Fugu Ultra v2 benchmark chart ↗

    Leader: Claude Opus 5.5 at 61.4%

  • GPQA Diamond#2 of 142
    95.5%vendorSakana AI: Fugu Ultra v2 benchmark chart ↗

    Leader: GPT-6 Astra at 96.1%

Other published results

ResultScoreReported bySource
GDP.pdf34.3%vendorSakana AI: Fugu Ultra v2 benchmark chart ↗
Chartography48.3%vendorSakana AI: Fugu Ultra v2 benchmark chart ↗
SWEFish (Sakana internal)71.8%vendorSakana AI: Fugu Ultra v2 benchmark chart ↗
DeepSWE74.3%vendorSakana AI: Fugu Ultra v2 benchmark chart ↗
ProgramBench81%vendorSakana AI: Fugu Ultra v2 benchmark chart ↗
Toolathlon80.6%vendorSakana AI: Fugu Ultra v2 benchmark chart ↗

Specs

Released
September 11, 2026
Parameters
Undisclosed
Size class
Undisclosed
Input
text, image
Output
text
Context
Not published
Max output
Not published
Cached input
$0.50 / 1M
Blended 70/30
$12.50 / 1M
  • Price: Sakana AI console pricing ↗ Up to 272K context; above 272K: $10 input / $45 output / $1.00 cached. Thinking tokens billed at output rate.

Will it fit your budget?

The planner prices this model against local hardware and other APIs for your workload.

Plan a stack →

Compare with

  • Sakana AIFugu Max
  • Sakana AIFugu
  • OpenAIGPT-6 Astra52.7
  • OpenAIGPT-6 Sol47.5