# Sakana AI

**Canonical:** https://apis.io/providers/sakana-ai/  
**Website:** https://sakana.ai/  
**APIs profiled:** 1

Sakana AI is a Tokyo-based AI R&D company founded by ex-Google researchers focused on collective intelligence, evolutionary optimization, and nature-inspired AI methods. The company's first commercial product is Sakana Fugu - a multi-agent orchestration system with an OpenAI-compatible API (currently in beta). Sakana also publishes Sakana Chat (consumer) and Sakana Marlin (beta).

## Kin Score — 10.9 / 100 (minimal)

Scored 2026-08-20 under rubric 0.12.0. Trend: flat (+0.0 from 10.9).

| Facet | Score |
|---|---|
| Discoverability | 68.5 |
| Contract Quality | 0.0 |
| Governance | 0.0 |
| Contract Governance | 0.0 |
| Operational Transparency | 10.5 |
| Developer Ergonomics | 0.0 |
| Commercial Clarity | 13.2 |
| Access Clarity | 13.2 |

## Agent readiness — 3.0 (human-only)

| Dimension | Value |
|---|---|
| Spec Presence | no |
| Agentic Access | no |
| Reversibility Documented | no |
| MCP Server | no |
| Auth Clarity | no |
| Idempotency | no |
| Error Semantics | no |
| OpenAPI Examples | no |
| Rate Limit Signal | documented |
| Event Surface Described | no |
| Agent Skills | no |
| Well Known Catalog | no |
| Consent Identity | no |
| Agent Card | no |
| Dry Run Mode | no |

## Access

Free — onboarding: unknown, pricing: free, trial: no (confidence: medium).

## APIs (1)

- **Sakana Fugu API (Beta)** — Multi-agent orchestration system that routes tasks across frontier foundation models (OpenAI, Anthropic, Google). OpenAI-compatible API surface. Two variants - Fugu Mini (low-la...

## Security (2)

- **Sakana Ai Domain Security** — TLSv1.3 · DMARC
- **Sakana Ai Vulnerability Disclosure** — security.txt · contact published

## Plans (1)

- **Sakana Ai Plans Pricing**

## Tags

Artificial Intelligence, LLM, Research, Foundation Models, Multi-Agent, Orchestration

---

Profiled by [API Evangelist](https://apievangelist.com) and published on [APIs.io](https://apis.io/providers/sakana-ai/). Scores are computed from the provider's own public artifacts under a published rubric.