All AI labs

Sakana AI

2 ranked models in a catalog of 3 text models. Fugu Ultra leads this lab at #71 overall.

Updated Oct 9, 2026, 06:15 UTCSI Score methodology

Ranked models

Global ranks · current snapshot
RankModelSI ScorecodingmathreasoningpreferenceConfidenceBlended priceContextReleased
#71Fugu Ultra
57.9
77.9—65.2—100%—1MJun 15, 2026
#80Fugu
55.9
69.6—63.3—100%—1MJun 15, 2026

Pillars use a 0–100 scale. * marks an imputed neutral prior where the model has no scored results in that pillar; it is not a measured benchmark result. Prices are USD per 1M tokens; see the blend and price comparison.

Scores and releases

4070100 Fugu: SI 55.9; released Jun 15, 2026; rank #80Fugu Ultra: SI 57.9; released Jun 15, 2026; rank #71Sakana Namazu: SI 53.5; released Aug 3, 2026; provisional Jun 15, 2026 Aug 3, 2026 4070100 Fugu: SI 55.9; released Jun 15, 2026; rank #80Fugu Ultra: SI 57.9; released Jun 15, 2026; rank #71Sakana Namazu: SI 53.5; released Aug 3, 2026; provisional Jun 15, 2026 Aug 3, 2026
Current SI Scores by release date, rather than historical scores. Filled dots are ranked; hollow dots are provisional. Month-only release dates sit at the start of their reported month.

Provisional models

1 awaiting rank-eligible evidence
  • Sakana Namazu

    Results cover 1 pillar; at least 2 required

    SI 53.5Aug 3, 2026

Latest Sakana AI news

All AI news ↗

What changed

All releases · RSS feed

Alerts on this device

What to be alerted about
RSS feed