Sovereign Intel, Localized Edge.
Exploring the limits of visual reinforcement learning, adaptive EFR routing architectures, and high-frequency edge LLMs built to run on localized consumer silicon.
Hayai 8B & 4B: Re-architecting Edge Vision for Mass Automation
Introducing EFR (Evolutionary Functional Reconstruction) - next-generation visual models that reinvent their weights during inference.
The Zeupiter Research Lab is preparing for the release of an incoming research paper detailing our next-generation visual foundation models: Hayai 4B & 8B (早い - meaning fast in Japanese).
Running full visual screen analysis at 30-second intervals can be cost-prohibitive on public endpoints. Hayai is our solution: a suite of visual LLMs designed to run locally on consumer-grade edge GPUs while maintaining performance parity with models 5x their size.
Evolutionary Functional Reconstruction (EFR)
Unlike traditional models with static parameters, Hayai is built on our proposed EFR architecture. During inference, the model dynamically reconstructs its active weight paths and utilizes adaptive routing blocks based on the complexity of the visual UI it receives. Simply put: the model functionally
By dynamically allocating weights, a Hayai 8B model can reduce active parameter usage down to 1.8B for simple visual tasks (like finding a login button) and scale up to its full capacity for dense, text-heavy spreadsheets.
Breaking the Transformer Constraint
To achieve high-frequency visual loops, we constructed a novel hybrid architecture designed to directly compete with Transformer-based models and Mamba/State Space Models (SSM). This hybrid approach eliminates the quadratic complexity of standard attention mechanisms, allowing the model to handle rapid screenshot streaming with flat performance latency.
Technical Capabilities:
- Latency: Targeting sub-20ms visual inference.
- Throughput: Over 5,000 tokens/sec on consumer-grade GPUs (e.g., RTX 4090/3090).
- Footprint Expansion: Engineered specifically to support autonomous, selector-free navigation across more than 100 million websites.
An update from Zeupiter: The Road to Sight-Driven Autonomy
A zoomed-out pivot from DevOps cloud infrastructure to scaling visual AI workforces across the global internet.
We are currently a small, highly aligned team of 5 people. Right now, every line of code represents a conscious choice about how we want intelligent software to interact with the world.
As a quick company update, Zeupiter has recently undergone a significant zoomed-out pivot. We originally started by solving DevOps and cloud infrastructure challenges. However, as we engineered those lower-layer routing nodes, we realized the underlying sandboxing and browser orchestration technology had massive applications far beyond DevOps workflows.
The Opportunity of Sight-Driven Autonomy
Rather than relying on fragile DOM selectors, CSS coordinates, or browser-specific XPath automations (which fail the moment a developer changes a class name), our core execution engine operates through pure visual understanding of the viewport. This allows our Sight-Driven Agents (SDAs) to navigate, reason, and execute complex workflows across the internet exactly the way a human operator would.
The pivot expanded our addressable market from a niche audience of systems engineers to a much broader category of internet users, developers, marketers, government bodies, and businesses looking to automate complex digital workflows. Since launching SDAs, we have already started seeing strong early user adoption, organic revenue generation, and product validation that reinforces our deepest technical convictions.
Key Architectural Milestones (Roadmap 2026):
- Fear Recall Intelligence & KTO-inspired RL: Our agents utilize an experimental, visual Kahneman-Tversky Optimization (KTO) reinforcement learning model. They learn from past interface failures ("fear recall"), preventing loop traps and navigating anti-bot screens with self-healing decision steps.
- Execution Across 20 Million Websites: Scaling visual thread clusters to safely map, isolate, and orchestrate automated tasks across a broad footprint of legacy portals and SaaS pages.
- Coding Agents: Integrating dedicated browser sandboxed IDE & harnesses that compile, run, and self-correct software tasks, outperforming standard static coding harnesses.
- Sub-20ms Latency & 1,000 tokens/sec on Consumer GPUs: Porting lightweight visual weights directly onto localized edge containers.