Fugu Max: benchmark results
Sakana AI's Fugu Max (v1.0), a multi-agent orchestration system served as a single model through Sakana's OpenAI-compatible API, released September 11, 2026. No downloadable weights. Provider: Other. Released 2026-09-11. Access: API.
Aggregate Rating 1780 ± 46, rank #126 of 10524 rated models, from 21 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Tinybird AI SQL Benchmark - First-Attempt Success Rate | 100 | Questions answered with a valid query on the first attempt ( | 94.5 |
| SvelteBench | 100 | Average pass@1 (%) | 93.7 |
| ObviousBench | 99.31 | Answer pass³ (%) | 89.9 |
| Enkrypt AI - Insecure Code Risk | 8 | Risk Score | 84.1 |
| Tinybird AI SQL Benchmark - Exactness | 52.69 | Result exactness vs human reference queries (0-100) | 74.7 |
| Tinybird AI SQL Benchmark - Success Rate | 100 | Questions answered with a valid query within 3 retries (%) | 71.9 |
| LLM Agronomy Benchmark (FBN) - Image Weed | 72 | Image Identification Accuracy (%, 50 questions) | 69.1 |
| Enkrypt AI - Risk Score | 27.5 | Risk Score | 67.6 |
| Enkrypt AI - Bias Risk | 78.81 | Risk Score | 65.2 |
| LLM Agronomy Benchmark (FBN) - V1 Questions | 86.7 | Category Accuracy (%) | 61 |
| LLM Agronomy Benchmark (FBN) - Soil and Water | 91.2 | Category Accuracy (%) | 50 |
| LLM Agronomy Benchmark (FBN) - Image Pest | 58 | Image Identification Accuracy (%, 50 questions) | 48.7 |
Interactive version: theaggregate.ai/model?slug=fugu-max · How It Works · Data refreshed daily, snapshot 2026-11-22.