Japanese AI startup Sakana AI has released Fugu, a multi-agent system that beats Anthropic's Fable 5 on multiple reasoning and coding benchmarks.
Sakana AI's Fugu orchestrates multiple models dynamically through a single API. In authoritative benchmarks across academia, reasoning, and programming, the top-tier Fugu Ultra scored higher than Anthropic's flagship models Fable 5 and Mythos Preview. Notably, Fable 5 and other controlled models are not part of Fugu's underlying model pool. Fugu achieves its results purely by orchestrating models like GPT-5.5, Gemini…