Home Solutions Product Research Contact Request Discovery Briefing
Energy-Based Model Architecture

Ulysses

The sensory nervous system for AI. Real-time intelligence on the hardware you already own.

Ulysses is a real-time sensory AI system built on Energy-Based Models, co-designed with Intel Labs. It processes raw, streaming, physical-world data at sub-10ns latency on standard Xeon CPUs — no GPU required.

Energy-Based Models Zero-Copy Inference Intel Xeon + AMX Sub-10ns Latency No GPU Required
Request a Demo → See How It Works
Explore

The Feed System

Ulysses sits between raw reality and every AI system downstream. It processes, detects, and learns from streaming data — then feeds clean intelligence to everything else.

Raw Reality

Streaming physical-world data arriving in real time from diverse sources.

  • Sensor arrays & IoT telemetry
  • Market data & financial feeds
  • RF signals & waveforms
  • Industrial process streams
  • Network traffic patterns
Stream

Ulysses

The sensory layer. Processes, detects, and learns from streaming data in real time.

  • Energy-Based Model inference
  • Zero-copy active inference
  • Real-time anomaly detection
  • Regime change identification
  • Continuous online learning
Feed

Downstream AI

Every model, pipeline, and system that needs to understand the physical world.

  • Large Language Models
  • ML prediction pipelines
  • Analytics dashboards
  • Domain-specific models
  • Decision automation

The data pipeline is simple: Raw Reality → Ulysses → Everything Else. Ulysses doesn't replace your AI stack. It makes every part of it better by giving it real-time sensory intelligence.

Why Not Transformers?

Transformers are brilliant at words. They're terrible at streaming physical-world data. AI needs both halves of the brain — and right now, one half is missing.

The Kingdom of Words

Cognitive System

Transformers and LLMs excel at language, reasoning, and structured knowledge. A $6 trillion ecosystem of proven capability.

  • Language understanding and generation
  • Reasoning and chain-of-thought
  • Knowledge retrieval and synthesis
  • Conversational interfaces
  • Code generation and analysis
  • Document processing at scale
The Kingdom of Numbers

Sensory System

Energy-Based Models process raw, streaming, numerical data from the physical world. This is where Ulysses operates — the missing half of intelligent AI.

  • Continuous streaming signal processing
  • Real-time anomaly detection
  • Regime change identification
  • Physical-world pattern recognition
  • Sub-millisecond inference on raw data
  • Online learning without pretraining

Ulysses doesn't compete with transformers. It completes them. The human brain has both a cognitive system and a sensory system. Current AI only has the cognitive half. Ulysses is the sensory system that feeds everything else.

Built Different at Every Layer

Not a better model on commodity infrastructure. A system where model and hardware execution are inseparable.

Zero-Copy Inference

Model and hardware co-designed as an inseparable system with Intel Labs. Zero-copy active inference architecture eliminates data movement overhead entirely.

Real-Time Learning

No pretraining required. Ulysses learns from streaming data as it arrives, continuously adapting to new patterns and regime changes without retraining cycles.

Edge-Native

Runs on Intel Xeon CPUs with AMX acceleration. Deploys where your data lives — at the edge, on existing hardware, with no specialized accelerators required.

Parameter Efficient

34x fewer parameters than comparable architectures. Energy-Based Models achieve superior performance with radically less computational overhead.

Sub-10ns Latency

Inference at the speed of hardware, not software. The zero-copy architecture delivers results in under 10 nanoseconds — fast enough for any real-time application.

Upstream Integration

Acts as the sensory front-end for any downstream system. Feeds LLMs, ML pipelines, analytics platforms, and domain-specific models with real-time intelligence.

Numbers That Rewrite the Rules

Benchmarked against comparable architectures on real-world detection tasks. These aren't theoretical — they're measured.

34x
Parameter Efficiency

34 times fewer parameters than comparable architectures while achieving superior detection accuracy.

100%
Regime Recall

100% regime recall on COVID detection benchmark. Standard models achieved 26% on the same task.

<10ns
Inference Latency

Sub-10 nanosecond inference. Hardware-speed execution through zero-copy active inference architecture.

0
GPUs Required

Runs entirely on Intel Xeon CPUs with AMX. Deploys on existing infrastructure with zero additional hardware cost.

Today: The Sensory Layer.
Tomorrow: The Sensory Infrastructure.

From vertical applications to horizontal platform. The roadmap from product to infrastructure.

Today — Vertical

The Sensory Layer

Ulysses delivers immediate value through targeted applications in signal processing and anomaly detection across specific domains.

  • Real-time signal processing on streaming data
  • Anomaly detection for financial markets
  • Regime change identification in industrial systems
  • Edge-deployed sensory intelligence for IoT
  • Domain-specific detection and classification
Tomorrow — Horizontal

The Sensory Infrastructure

Ulysses becomes the sensory front-end for all hybrid AI architectures — a universal layer between the physical world and every downstream AI system.

  • Universal sensory layer for hybrid AI stacks
  • Standard interface between reality and AI
  • Infrastructure-grade reliability and performance
  • Multi-modal streaming data fusion
  • Platform APIs for any downstream consumer
Vertical applications → Horizontal infrastructure platform

Co-Designed with Intel Labs

Optimized for Xeon with AMX. The model and hardware aren't just compatible — they're co-designed as an inseparable execution system.

Most AI companies build models and then find hardware to run them. Kavara did the opposite. Ulysses was co-designed with Intel Labs from the ground up, creating a zero-copy active inference architecture that treats the model and the silicon as a single system.

The result is inference at the speed of hardware, not software. No data movement. No memory copies. No GPU tax. Just raw execution on the billions of AI-capable transistors sitting idle in existing infrastructure — what we call dark flops.

Every Intel Xeon processor with AMX already has the capability to run Ulysses. That means the hardware is already deployed. It's already paid for. It's just waiting.

Intel Xeon + AMX

Native execution on Advanced Matrix Extensions without GPU dependencies.

Zero-Copy Architecture

No data movement overhead. Model execution fused with hardware memory paths.

Dark Flops Unlocked

Billions of AI-capable transistors in existing infrastructure, activated.

Intel Xeon
AMX
Zero-Copy Runtime

See it on your data.
On your hardware.

Ulysses deploys on the infrastructure you already own. No GPUs to procure. No cloud dependencies. Real-time intelligence, running today.