Sakana AI Labs launches Sakana Fugu beta
Sakana AI Labs launched Sakana Fugu beta, a multi-agent orchestration system that dynamically coordinates frontier foundation models. It selects optimal agent combinations and roles while routing tasks across open and closed AI models. Fugu achieves state-of-the-art results on SWE-Pro, GPQA-D, and ALE-Bench benchmarks. Sakana AI Labs co-founder and CEO David Ha announced the beta release.
First post
We’ve been using Sakana Fugu internally for our own research and coding. Instead of relying on a single model, it dynamically orchestrates the best combination of open and closed models for any task. The future of AI is collective intelligence. Excited to open the beta release:
We’re launching the beta for our new commercial AI product: Sakana Fugu 🐡, a multi-agent orchestration system! Blog: https://sakana.ai/fugu-beta Fugu hits SOTA on SWE-Pro, GPQA-D, and ALE-Bench, and has been our internal secret weapon. It dynamically coordinates frontier models, autonomously selecting the optimal agent combinations and roles for each task. Available as an OpenAI-compatible API, you can seamlessly integrate Fugu into your existing workflows with minimal changes. 🐟 Fugu Mini: High-speed orchestration optimized for latency 🐡 Fugu Ultra: Full model pool utilization for deep, complex reasoning Apply for the beta test here: https://forms.gle/BtKkhc2CfLKk1dvNA
Why it matters
Sakana AI ranked an agent 21st out of over 1000 humans in a 2025 AtCoder contest using ALE-Bench.
Sakana AI developed Fugu based on its Trinity evolved large language model coordinator paper at ICLR 2026.
David Ha (Sakana AI CEO) published a 2022 survey on collective intelligence methods for deep learning.
