Introducing Multi-Agent Self-Supervision (MASS): Recursive self-improvement through collective intelligence Blog: pub.sakana.ai/mass/ As AI systems tackle increasingly open-ended research problems, human experts may struggle to assess results and guide further progress. The model itself may be the best available optimizer and evaluator. The challenge is obtaining useful self-supervision within that homogeneous loop, where a model risks reinforcing its own blind spots rather than correcting them. We ask whether a model can learn from the collective intelligence of its own team. We introduce Multi-Agent Self-Supervision (MASS). One shared model solves tasks through a team of virtual subagents, searches for better multi-agent workflows, then uses its own judgments to select the best ones. Training on the best team’s executions provides self-supervision by distilling the team’s collective experience into the shared model. We ran two cycles using a 27B open-weights model on synthetic open-ended research tasks. After the second cycle, score per output token reached 1.2-1.6x the base model's level across four research benchmarks. Three key results: First, MASS enables role generalization within an RSI loop: training only on task-solving data from the optimizee naturally improves the capabilities of both the optimizer and the evaluator. Second, multi-agent trajectories are more efficient training data than single-agent trajectories for RSI. Third, optimizer capacity, rather than evaluator capacity, can be the bottleneck in the speed of improvement. We believe the community needs to investigate homogeneous RSI carefully to support AI safety. When a model is also its own evaluator, shared blind spots could allow mistakes to be accepted as progress and reinforced through training. For AI to improve beyond human expertise safely, we need to understand these failures and preserve effective ways to detect them and intervene. Paper: arxiv.org/abs/2610.12176 Code: github.com/SakanaAI/mass This work was a collaboration with researchers at UC Berkeley.

Introducing "Scaling In-Context Imitation Learning" (SAIL) to be presented at #IROS2026. This work is a collaboration between Sakana AI and the University of Tokyo. Blog: pub.sakana.ai/sail What does a robot need before it can tackle a new task? Teaching a robot something new usually starts with collecting demonstrations and training a policy. But foundation models have already learned from vast amounts of images, text, and robotics-related data. We wanted to see how much of that knowledge we could draw out for robot control without changing the model itself. Recent demonstrations suggest that GPT-6 Astra can operate physical robots alongside its general language and vision capabilities. Earlier work has also shown that LLMs/VLMs can generate entire sequences of robot movements from a few demonstrations. However, a foundation model does not necessarily produce a reliable robot trajectory in a single generation. Performance depends on the context provided, and a small error in a movement target can cause the entire task to fail. We propose SAIL, a method for more reliable VLM-based robot trajectory generation through test-time scaling. SAIL uses a policy VLM as a robot trajectory generator, conditioned on a few successful demonstrations. It tests the generated trajectory in a simulator and uses an evaluation VLM to review the resulting video and identify where progress stalled. The policy VLM then uses this feedback to revise the trajectory, with Monte Carlo tree search (MCTS) exploring alternatives while refining promising candidates. Only the selected trajectory is sent to the physical robot. Across six manipulation tasks in simulation, increasing the search budget from one candidate to 45 raised the average rate of finding a successful trajectory from 25% to 73%. We also evaluated SAIL on a physical robot. Our results suggest that robot trajectory generation can benefit from test-time scaling, with additional computation enabling the model to test and refine its proposed actions in simulation. We think there is more to learn about what existing models can do with this kind of feedback, and how far those improvements carry over to physical robots. Paper: arxiv.org/abs/2603.08269 🐟

Introducing PC-ALM, a local-learning alternative to backpropagation. Our method trains 1000-layer neural nets using only local dynamics, and without backprop. Blog: pub.sakana.ai/pc-alm/ Standard deep learning relies on backpropagation. The brain, however, cannot implement backpropagation, at least not exactly. How can a physical system, such as the brain, solve multilayer credit assignment without explicit use of backprop? We look for inspiration in two related fields: distributed optimization and NeuroAI. In NeuroAI, predictive coding asks each neuron activation to solve an energy-based inference problem instead of using a standard forward pass. That inference step can be implemented as energy-minimization dynamics on local prediction errors. This perspective -- each layer as a dynamical system -- has proven promising, but performance of predictive coding hasn't scaled well with depth. Credit signals at far ends of the network struggle to diffuse into internal layers. We turn to distributed optimization, generalizing predictive coding to use an augmented Lagrangian instead of energy. This motivation stems back to a classic 1988 paper by LeCun, showing that the Lagrange multipliers of a deep network can be identified with gradients of a supervised loss. The augmented Lagrangian then bridges LeCun's perspective to the standard predictive coding that is used in NeuroAI. We find that this new perspective yields a natural PC-like alternative to backpropagation, resulting in a method we call PC-ALM. PC-ALM differs from PC in that it introduces dual neurons (Lagrange multipliers) as part of the layer-local dynamics, resulting in each layer acting as a PI feedback control system to minimize local prediction errors. We find that PC-ALM is capable of propagating signals to seemingly arbitrary depth, especially in deep narrow networks where standard PC struggles to learn. Ultimately, our motivation here is to understand how distributed physical systems, such as the brain, can compute credit signals using only local coupling and local dynamics. PC-ALM may also inform deep learning in neuromorphic hardware, where dynamics are cheaper than on GPUs. Paper: arxiv.org/abs/2605.31022 Code: github.com/SakanaAI/pc-alm

Introducing Fugu Max and Fugu Ultra v2: the next evolution of Sakana Fugu’s multi-agent orchestration system. Try: sakana.ai/fugu Blog: sakana.ai/fugu-max-release/ The frontier that actually matters is the Pareto frontier: capability on one axis, cost on the other. But the industry still treats it as a static menu of isolated models. Today we are resolving that with a dynamic architecture: Fugu Max expands the Pareto efficiency frontier. By orchestrating our largest pool of open-weights and specialized models to date, including NVIDIA Nemotron family, it dynamically routes tasks to the leanest capable model. Fugu Max delivers performance within striking distance of elite models at two to six times lower cost. Fugu Ultra v2 pushes the peak capability of orchestration higher than ever before. On Chartography, it outperforms Opus 5 and Fable 5. On DeepSWE, it outperforms models that cost three to five times more per token. Crucially, it does all of this without Fable 5, Fable 5.1, or GPT-6-Astra in its agent pool. The Fugu orchestration system evolved with model resiliency in mind. It does not rely on individual frontier models to deliver frontier output. By orchestrating a swappable pool of open and specialized models, it outperforms closed ecosystems while protecting users from vendor lock-in, API revocations, and sudden service cutoffs. Fugu Max expands the Pareto frontier outward. Fugu Ultra v2 pushes it upward. Orchestration does not force a choice between cost and capability. It advances both simultaneously.

Sakana AIは、ＳＣＳＫ株式会社および住友商事株式会社と、AI活用による日本の産業変革と社会課題解決に向けた包括業務提携を締結しました。 sakana.ai/scsk-sc-partnershi… Sakana AIの技術力、ＳＣＳＫの統合・実装力、そして住友商事の事業基盤を掛け合わせ、AIの社会実装を加速させます。

Sakana AIリサーチサイエンティスト・菅沼雅徳のインタビューが9月8日の朝日小学生新聞に掲載されました。 リサーチャーの仕事内容や1日の流れ、生成AIの未来、小学生のAI活用法についてお話ししました。 機会をいただいた編集部のみなさまに感謝します。

Sakana AIは、今月より一般社団法人AIロボット協会（AIRoA）の正会員になりました。 RSI Labの取り組みの一環として、@airoa_jpと協働しながら、世界モデルやPhysical AIの研究開発を進めていきます。 airoa.org/ja/members-ja/ 🐟

🐟 Sakana Namazu API 公開 🐟 本日、Sakana AIは大規模言語モデル「Namazu」をアップデートし、API「Sakana Namazu（サカナ・ナマズ）」として提供を開始しました。 Sakana Namazu API: sakana.ai/namazu 🐟

Announcing Fugu-Ultra v1.1 🐡 We’ve been thrilled by the reception to the Fugu model family. Thanks to everyone who tried it, shared feedback, and trusted Fugu with real work. Today, we’re releasing Fugu-Ultra v1.1 → sakana.ai/fugu Upgraded to incorporate the latest frontier models, resulting in stronger performance across every benchmark shown, including gains of up to 7.9 points over v1.0, with particularly strong results on ProgramBench and Terminal Bench 2.1. Fugu-Ultra v1.1 is more capable across coding, agentic tasks, and advanced reasoning, and available at the same price as Fugu-Ultra v1.0 The frontier keeps moving, and Fugu keeps getting better.

Introducing Fugu-Cyber: an update to our Fugu orchestration model. It achieves state-of-the-art performance on real-world security benchmarks, matching cyber-focused frontier models like GPT-5.5-Cyber and Mythos Preview. sakana.ai/fugu-cyber-release 🐡

東京・八芳園で開催された「NVIDIA Japan AI Ecosystem Reception」に、Sakana AI がブースを出展しました。 多くの方に足を運んでいただき、私たちの技術やプロダクトを直接お伝えする貴重な機会となりました。 お声がけくださった @NVIDIAJapan に、心より感謝申し上げます。

数週間分の戦略リサーチを、数時間で。 あなたのVirtual CSOとして働く、Sakana Marlin。 最初のテーマを渡してみる： sakana.ai/marlin 🐟

We’re Hiring: Software Engineer (R&D, Infrastructure and Platform Reliability) 🐟 sakana.ai/careers/infrastruc… Sakana Fugu, our Multi-Agent System as a Model, launched publicly last month and is growing fast. We’re hiring an engineer to keep Fugu fast, dependable, and cost-efficient at scale. ・GKE, Vertex AI & Terraform on GCP ・AWS services equivalent to GCP services are also appreciated ・Monitoring, incident response & on-call ・High autonomy, working closely with R&D and Product Based in Tokyo. If this sounds like you, we’d love to hear from you 🚀